Computing Service Architecture: central monitor-absence load balancing

Size: px
Start display at page:

Download "Computing Service Architecture: central monitor-absence load balancing"

Transcription

1 Computing Service Architecture: central monitor-absence load balancing Satoshi Azuchi, Sojeong Hong, Jinoh Kim May 5, 26 Abstract We propose decentralized load balancing by priority and migration based load control mechanism, which is done autonomously. Each computing server s load balancing is achieved by dynamically changing its priority and control task acceptance rate based on a computing server s priority. In addition by providing task set level migration, we can enhance the effectiveness of our decentralized load balancing mechanism. For security, we provide mutual authentication and capability based access control. We implemented a mock of computing service architecture and support our claims by presenting several sets of experiment results in this report. 1 Introduction With tremendous growth of personal computers, distributed computing has proliferated. The distributed computers enable to solve complex tasks such as scientific computation without having an expensive super computer. The typical examples of the distributed computing are Grid and volunteer computing based on donation of computer cycles such as SETI@home [2]. The volunteer computing assumes heterogeneous computing environment due to the characteristic of voluntary participation of the computing resources, while the Grid is built upon homogeneous controlled environment. While deploying distributed computing, even though it exploits a lot of CPU cycles, some problems have been announced as well. One of the problems is that distributed computing systems are vulnerable to single point of failure, especially monitor server which is responsible to manage and control the distributed computing resources. Employing additional computing hardwares for duplication can solve this problem, but it requires extra costs. Moreover, it does not achieve only by replicating the monitor server; rather it requires complex and accurate fault tolerant protocols and mechanisms. Hence, we propose a new distributed priority-based computing architecture to utilize the computing resource more efficiently and be tolerant against single point of failure. The traditional distributed computing systems do not much consider the priority-based computing services. There may be some tasks simultaneously, and each of them requires different set of resources (in this paper, we use job and task interchangeably). Therefore, the priority-based computing services can be applied for resource management and allocation in distributed computing environment. This functionality is used to apply policies for resource utilization in some sense as well. Security concerns are also an important part of the distributed computing. In particular, donationbased computing has more potential problems than the Grid computing. As mentioned above, an institution organize the Grid, so it is simple to protect the resources. Contrarily, it is difficult to protect resources in the voluntary computing environment because there is lack of well-organized management functions. This arises some security issues in distributed computing in terms of network attacks as well as authentication and authorization. 1

2 In order to solve these problems, we consider the central monitor-absence architecture; that is, fully distributed. By removing the monitor server, we can save cost and reinforce fault-tolerant capability because we do not worry about the master server down. In addition, by providing priority-based computing services, we can offer users to have different degrees of computer cycles. However, removing the master server requires another load balancing method because the master server is in charge of load balancing. For this reason, we propose a central monitor-absence load balancing method for our architecture. 2 Computing Server Farm Design In this section we elaborate our computing server farm design. First, we introduce overall architecture. Then a scenario is described and each component is explained precisely. 2.1 Architecture The proposed architecture provides virtual machine-based load balancing without any central load balancing monitor, capability-based access control, and mutual authentication. The overall architecture consists of computing server farm, service management server, and clients as shown in Figure 1. Each component s role is explained in the following sub-sections. Service Management Server 1. Auth. 2. Subscription 3. Capability Computing Server Farm 3. Auth. 4. Job submission 5. Result Client Figure 1: Design overview 2.2 Scenario Before using the computing service, the user should buy/receive a degree of computer cycle, so called subscription or service level agreement. Based on the client s payment or client s role (if the priority is decided based on the user s role in a company), the priority is decided by the service management server. The service management server makes capability and signs the capability. Then it sends the capability token to the client. 2

3 When a client wants to run his job on the computing server farm, it requests for the availability of the server usage with capability token; the server selection can be done randomly (we assume that a client knows all server s address). When the server receives the request, it checks the capability s validity such as expiry date and signature of service management server. Then it compares the received client s priority and the machine s priority. If the client s priority is higher or equal to the machine s priority, it accepts the request and asks the client to submit the job. Otherwise, it forwards the request to another machine. In the computing server farm side, each machine maintains its priority and changes the priority dynamically based on its workload. Specifically, if a server has more jobs than a certain number it increases its priority so that it can receive jobs from a fewer clients who have higher priority. The rejected client can contact other computing servers iteratively until it can submit its job. Our design supports an additional load balancing via task set migration across computing servers (due to the implementation decision, we provide task set level migration using Xen s migration function instead of single task migration, process migration). The task set migration takes places as follows. A computing server has a threshold, which is the maximum number of tasks a computing server can hold. An administrator can decide the threshold based on their resource capability (e.g. CPU speed, size of memory, etc). When the number of jobs reaches to the threshold, the computing server stops accepting clients task submissions. Then it communicates with other computing servers and migrates a task set to another computing machine, whose workload is the least. If all computing server is too overloaded to accept task set migration, a computing server wait until some of its tasks end and start to accept clients task submissions again; the client with the highest priority will get this chance. 2.3 Major Components Computing Server Load Balancer : A server farm consists of a set of autonomous load balancing enabled computing servers. Each computing server maintains priority variable P, which is dynamically adjusted based on each server s workload. This work is done by the load balancer. The load balancer s responsibilities are listed as follows : Work distribution inside a machine : when a load balancer receives a task from a client, it sends to computing daemon on top of virtual machine. Priority Adjustment : A load balancer has pre-defined work unit t. If the workload is greater than t, it increases P by one. If the workload gradually increases and reaches to 2t, P is increased by one again and so forth. Job Migration Decision : A load balancer monitors the total number of tasks it holds. When it reaches to the threshold, which pre-defined by the administrator based on the server s capability, the load balancer negotiate with other computing servers and trigger the task set migration; the load balancer selects the computing server whose workload is the least. Computing Daemon : The accepted task is actually executed by the computing daemon on a virtual machine. Access Guard : Each computing server also has an access guard. The access guard verifies the expiry date and the digital signature [9] of the service management server in the submitted capability to validate the capability s integrity and authenticity. Then the access guard checks if the priority in the capability is greater or equal to the computing machine s current priority P. If the client s priority is less than P, it forwards the request to another computing server. Otherwise, it accepts the client s request and let the load balancer accept the job. 3

4 2.3.2 Service Management Server The service management server is the entity that issues capability to the client. The capability includes 1) identity of a client, 2) priority, 3) expiration time, 4) the service management server s digital signature. The priority of a client can be decided based on how much the client pays for the computing service subscription. Note that the service management server does not interact with any computing server in the farm. Hence, as long as a client has a valid capability token, it can use computing server farm even if the service management server is off-line Client The client is the entity that submits jobs to the computing server farm. Before any job submission, it should receive a capability from the service management server. When a client wants to submit a job to the computing server farm, it requests job submission to a server randomly. Note that at the first request, the client does not submit the job actually. If the computing server has lower priority and it accepts the client s request, the client actually submits the job. 3 Implementation All components are implemented using C and for experiment and result parsing, we used Perl. Computing servers are installed on three machines in the itlabs instructional cluster. The service management server and a set of clients (processes called by Perl script) are installed on tera.cs and oxygen.cs respectively. 3.1 Xen installation and virtual machine migration We installed Xen 3. onto three Ubuntu 5.1 machines in the itlab instructional cluster. Xen installation process automatically compiled Xen and Linux kernel for Xen (called XenoLinux) based on kernel This XenoLinux always runs on Xen in order to provide control interface of Xen. Then, we confirmed that we were able to run virtual machines on Xen and to migrate them from the one machine to another. We used Debian (sarge) for this testing. To support OS migration, a root disk image must be shared among machines. First, we tried to use the global network block device (GNBD). However, it didn t work on our environment. Thus, we now use NFS to share a root disk image. 3.2 Computing server We implemented and installed load balancer, which assigns jobs to virtual machines and manage all jobs on a computing server, on three machines (on Domain ) in instructional cluster. Each machine has four virtual machines (Ubuntu 5.1) and computing daemon is running on top of the virtual machines to compile and run submitted tasks. When the task ends, it reports to the hosting load balancer. When a migration takes place, the load balancer transfers the information about the migrated task set and update its database. Note that the computing daemon, which has each task s owner information (e.g. IP address), is also migrated. Hence, when a task is finished, the daemon can notify to the new load balancer on the new machine. 3.3 Multi-threaded TCP/IP communication We implemented a communication module to provide middleware functions based on TCP/IP sockets among clients, the service management server, computing servers, and virtual machines. Each component has a server daemon to receive any service requests from other components. To request services of any other components, transient connections are established because permanent connections can be 4

5 limited by system configurations. After communicating, the communication channels are closed. In particular, since there can be a lot of clients, transient connection management is better in terms of scalability. As the architecture, the communication module is designed using multi-threaded event-driven model. Thus each component waits for an event, and a thread is spawned whenever an event comes. Then corresponding callback function is called to provide appropriate service. By providing well-defined APIs, the module works as a middleware in the system. Table shows the APIs provided by the communication module to applications. API Name comm. nonce comm. subscr client comm. submit client comm. forward task comm. send result comm. nego migrate Description Exchange mutual authentication data Request/reply subscription of clients Request/reply submission of tasks Forward tasks to a VM Return the result of a task Exchange VM migration information Table 1: Communication APIs 3.4 Public key infrastructure based authentication We implemented standard authentication protocol based on public key infrastructure. First, for public key authentication, we used X.59 [12] package provided by OpenSSL; X.59 specifies the standard for standard formats for public key certificates and a certification hierarchies. Service management server is considered as root certificate authority (which has a certificate signed by its own private key and issues certificates to other entity) and a key generator (it generates public key pair for computing servers). In our system, the service management server issues certificate for computing servers and clients. In our system, we do not consider connection hijacking so that when the client could verify the server s IP address, we consider it as the servers are authenticated. To authenticate a server, a client sends a nonce (random number) to the server, and then the server is supposed to sign the nonce. The server sends the signed nonce and its public key certificate. We assume that all clients have the root certificate authority s certificate already. When the client verifies the signature, it sends its data to the server, otherwise it conclude the server is not valid one. To protect the capability content, especially to prevent for the client to modify its priority, service management server sign the capability content and the computing servers checks the validity of capability using server s public certificate. This signature generation and verification is implemented using PKCS#7 [13] package, provided by OpenSSL. 4 Experiment Results In this section, we show comparison of computing servers workload and clients waiting/turnaround time based on task submission traffic with diverse properties. Also, we show the effect of task set migration to load balancing. The traffic is generated using Poisson random number generator. At each time interval (4 second), client processes are generated based on the random number. We control the rate of task submission using λ value. Also, we manipulate priority distribution among clients to show its effect on server; details are explained in the following sub-sections. 5

6 3 # of tasks legolas # of tasks gimli # of tasks aragorn 25 2 # of tasks sec Figure 2: Well-balanced computing servers workload 3 # of tasks legolas # of tasks gimli # of tasks aragorn 25 2 # of tasks Figure 3: Computing servers workload based on weighted random selection. aragon, gimli, and legolas are selected with.6,.3, and.1 probability as a first choice sec 6

7 4.1 Server workload with respect to diverse traffic For the first test case, the traffic with λ = 3 (45 job submissions per minute) is generated for 1 minutes. In this case, the service management server randomly selects the priorities of clients and a client also selects computing servers randomly to submit its task. Figure 2 shows that computing servers workload based on the given traffic. The three machines workloads are well balanced as we expected. After about 6 seconds, no traffic is not generated. However, the computing servers still have running tasks. That is because the remaining clients, whose priority was less than computing servers, are accepted by computing servers since they become less busy. Since with the first case traffic, it is not sufficient to show our priority based load balancing works properly. We generated different kinds of client set that select one computing server with.6 probability, i.e. this server (aragon) is considered a more popular computing server compared to the other two computing servers. Hence when a client starts to contact computing servers there are more chances that aragon will be selected. The other two computing servers are chosen with.3 and.1 probabilities (the corresponding servers are gimli and legolas respectively). If a client gets rejected from the computing server it choose with weighted random selection, the client contacts another computing server. Figure 3 shows the fairly well-balanced servers work load even if the clients chooses aragorn as a first attempt with higher probability. So far, we tested computing servers with the set of clients whose priorities are selected randomly between 1 and 5 (the lowest and the highest priority in our system) by service management server. What if the priories of clients are not well distributed? We generated two skewed distributions, clients with only high priorities (4 or 5) and clients with only low priorities (1 or 2). Figure 4 shows how workload changes on a computing server with respect to different priority set clients. The first workload peak is generated by the clients with evenly distributed priorities. The second workload peak is generated by the clients with 1 or 2 priorities and the last one is derived by the high priority clients. Except the different priority combination, other conditions (λ = 3, computing servers are selected randomly by clients) are set to the same values for the equality. Since the computing server control its workload based on priority, it could block the lower level clients task submission dynamically. However, in case of high priority client traffic, server could not reject clients requests a lot since clients priority were higher than computing servers. 4.2 Task set migration effect How does the task set migration affect on load balancing? If computing server can migrate tasks when it is overloaded, it is obvious that task migration is beneficial for load balancing. Here we will show the effect of task set migration by comparing workload on computing servers with virtual machine migration and without it. In order to trigger the migration, we generate a special client task submission pattern by giving them all priority 5. Clients select three computing servers with weighted probability of.6,.3, and.1 as we presented before. Note that we do not consider load balancing based on priority here to show only the effect of task set migration. Figure 5 shows the case of no migration. The workloads of three systems are the same as the load injected by 6:3:1 proportion. This is because the overloaded system denies submitted jobs when it reaches to the threshold and the clients choose another machine. In the case of migration in Figure 6, workloads of three systems are a little different from the injected loads. However, the overloaded system shows significantly shorter peak duration. This is very important. The overloaded system is aware of over-utilized situation, and it attempts migration immediately. Therefore it can be easily free from overloaded situation. Around 55 seconds, however, the system experiences relatively long overloaded period. This is because the system cannot migrate more virtual machines due to system configuration (the maximum number of virtual machine which can be migrated is set to 2 in the configuration in each system). For the results, it is clear that virtual machine migration is a great idea for load balancing. 7

8 25 pri 1/2 pri 4/5 all pri 2 15 # of tasks 1 5 sec Figure 4: Work load of a computing server with different clients priority combinations; 1-5 priority distribution, 1 or 2 priority distribution, 4 or 5 priority distribution from the left to the right 4.3 Comparison of clients with respect to priority We have seen that our priority and migration based load balancing works well. In this section we will show the efficiency of our system in the point of client. We measured clients waiting time, turnaround time and the number of rejections by computing servers. In order to manipulate the workload in computing servers, we generate two kinds of tasks (1 minute and 3 minute durations tasks) with same rate of task submissions. In order to submit task, a client communicates with computing servers iteratively, i.e. if a computing serveer rejects a client, the client communicates with other computing servers. If all servers reject a client s request, the client waits for 1 seconds and starts to pull the computing servers again. Figure 7 shows the average waiting time and turn around time with respect to clients priorities. The lower priority clients have waited longer time than the higher priority clients. In addition, when the computing servers are more overloaded (i.e. with 3 minute tasks), the waiting time become much longer than light workload on computing servers. Figure 8 shows the average number of rejections from computing servers with respect to clients priorities. The clients with priority 1 waited for about 85 times when the task duration was 1 minute and about 33 timers when the task duration was 3 minute, on average. The clients with priority 2 have rejected for very a few times when the task duration was 1 minute and about 13 times when the task duration was 3 minutes. The clients with priority 3 have rejected about 25 time only when the task duration was 3 minutes. The higher priority clients were not rejected during our experiment settings. The result is what we exactly expected. Hence, our design can provide differential service based on priorities. 8

9 3 # of tasks legolas # of tasks gimli # of tasks aragorn 25 2 # of tasks Figure 5: Workload on computing servers with weighted random selection without task set migration. The client has only 5 priorities. sec 3 (2) (3) # of tasks legolas # of tasks gimli # of tasks aragorn 25 (1) 2 (1) aragon -> gimli (2) aragon -> legolas (3) gimli -> legolas # of tasks Figure 6: Workload on computing servers with weighted random selection with task set migration. The client has only 5 priorities. sec 9

10 9 8 waiting time with 6s job waiting time with 18 s job turnaround time with 6s job turnaround time with 18s job 7 6 time (sec) 5 4 time (sec) Figure 7: The average clients waiting and turnaround time based on their priorities. 18-second tasks derive more workload on computing servers than 6-second tasks priority 35 # of rejects with 6s job # of rejects with 18 s job 3 25 # of rejects Figure 8: The average number of rejections by computing servers with respect to clients priorities. 18- second tasks derive more workload on computing servers than 6-second tasks priority 1

11 5 Related Work Network load balancing is a critical area to offer online services because users do not want the services are procrastinated or unresponsive. In particular, server load balancing is important since unavailability is one of the crucial measures the sites [4] [14]. There are several ways to provide load-balancing function for sites servers. Server clustering is a promising technique to build scalable server architecture [18]. Geographical load balancing is another way to achieve the distribute the load [5]. It distributes the load using network proximity information, so the closest site is selected for the service, while the server clustering is usually take the CPU load of each system into account to distributed the load. Grid computing enables users to obtain higher throughput computing by taking advantage of many networked computers. Especially, it uses the CPU resources of individual computers which are connected networks to solve large-scale computation problems [7] [15]. Volunteer computing is a new paradigm: it uses donated computing resources, while grid computing uses well-organized recourses. The typical example is SETI@home [2] run on BOINC distributed computing infrastructure [1]. Experimental work to build grid computing environment using Xen [3] for scientific computing was provided by [11]. In this work several virtual machines are installed on a machine to provide grid computing environment. In addition, PVM [8], which virtualizes a number of heterogeneous machines as a single distributed parallel processor, is a famous tool for scientific and engineering computing. Virtual Machine migration mechanisms have been developed because encapsulating the state of a running process which is necessary for the process migration is difficult [6] [17]. [6] integrated live OS migration into Xen which enables rapid movement of interactive workloads within clusters and data centers. [6] implemented VM migration system called VMotion which is a part of the VMware Virtual Center Product. 6 Lessons learned Priority based load balancing enables us to eliminate a central monitor since all servers achieve self-load balancing autonomously. However, one problem takes place; a client who has low priority cannot submit his job to any of computing servers if all servers are busy, i.e. the priorities of all computing servers are higher than that of the client. This problem can be alleviated 1) by allocating a certain number of clients that a computing server would accept even if the server has higher priority, or 2) by having a dedicated server for low priority clients. In the case of the first approach, we can also employ a tunable policy that the number of acceptable lower clients can be changed by the server administrator s decision. More precisely, if the number of acceptable lower clients is, the system does not consider the unlimited waiting of the low priority client. In the opposite way, if the value is the number of possible jobs a server can hold, the server does not follow our priority based load balancing. Hence, the value should be determined heuristically. Task set migration might not be the right choice to show how task migration among computing servers would be beneficial for the load balancing. We could use mobile processes (such as Aglet [16] or D Agents [1]) for single task level migration. To balance load, we only took the iterative contact to servers into account. The iterative way works pretty good in our experiments, but we are not sure that there might be same results if we use more complicated or refined variables (e.g., different processing time for each job). With respect to this concern, the recursive contact can be a good approach, in which the servers forward the job by negotiations. 7 Conclusion The traditional way of load balancing based on centralized monitor is vulnerable to single point of failure. If we multiplicate the monitor, then the cost increase is increased not only due to the monitor itself cost 11

12 but also due to the management cost. Hence, we proposed priority and migration based decentralized load balancing. Priority based autonomous load balancing provide load balancing and differential service based on how much a client pays for the service. Experiments show that even if there is skewed preference among computing servers, our system could achieve load balancing. Furthermore by providing task set migration, we also enhance the effectiveness of our load balancing mechanism by moving task from the overloaded server to the unloaded server. Last, capability-based on access control make computing servers independent from the service management server. Mutual authentication can protect clients data from attacker who might impersonate computing/service management server and computing server from malicious users. References [1] D. P. Anderson. BOINC: A system for public-resource computing and storage. In GRID, pages 4 1, 24. [2] D. P. Anderson, J. Cobb, E. Korpela, M. Lebofsky, and D. Werthimer. Seti@home: an experiment in public-resource computing. Commun. ACM, 45(11):56 61, 22. [3] P. Barham, B. Dragovic, K. Fraser, S. Hand, T. Harris, A. Ho, R. Neugebauer, I. Pratt, and A. Warfield. Xen and the art of virtualization. In Proceedings of the nineteenth ACM symposium on Operating systems principles, volume 37, 5 of Operating Systems Review, pages , New York, Oct ACM Press. [4] T. Bourke. Server Load Balancing. O Reilly, 5th edition, 21. [5] V. Cardellini, M. Colajanni, and P. S. Yu. Geographic load balancing for scalable distributed web systems. In MASCOTS, pages 2 27, 2. [6] C. Clark, K. Fraser, S. Hand, and J. Hansen. Live migration of virtual machines. In In proceedings of the 2nd Symposium on Networked Sytems Design and Implementation, 25. [7] I. Foster and C. Kesselman, editors. The Grid: Blueprint for a New Computing Infrastructure. Morgan Kaufmann, San Francisco, CA, [8] A. Geist, A. Beguelin, J. Dongarra, W. Jiang, R. Manchek, and V. Sunderam. PVM: Parallel virtual machine: a users guide and tutorial for networked parallel computing. MIT Press, Cambridge, MA, USA, [9] Goldwasser, Micali, and Rivest. A digital signature scheme secure against adaptive chosenmessage attacks. SICOMP: SIAM Journal on Computing, 17, [1] R. Gray, G. Cybenko, D. Kotz, and R. Peterson. and d. rus. d agents: Applications and performance of a mobile-agent system, 21. [11] M. Hardt and R. Berlich. Xen: Scientific use cases and performance comparisons, 25. [12] R. Housley, W. Ford, W. Polk, and D. Solo. RFC 2459: Internet X.59 public key infrastructure certificate and CRL profile, Jan Status: PROPOSED STANDARD. [13] B. Kaliski. Pkcs #7: Cryptographic message syntax, version 1.5. RFC 2315, March [14] C. Kopparapu. Load Balancing Servers, Firewalls, and Caches. Wiley, 22. [15] K. Krauter, R. Buyya, and M. Maheswaran. A taxonomy and survey of grid resource management systems for distributed computing. Softw, Pract. Exper, 32(2): , 22. [16] D. B. Lange and M. Oshima. Mobile agents with Java: The aglet API. World Wide Web Journal, [17] M. Nelson, B.-H. Lim, and G. Hutchins. Fast transparent migration for virtual machines. In In proceedings of Usenix 25 Annual Technical Conference, pages ,

13 [18] T. Schroeder, S. Goddard, and B. Ramamurthy. Scalable web server clustering technologies. Network, IEEE, 14(3):38 45, May 2. 13

Xen Live Migration. Networks and Distributed Systems Seminar, 24 April 2006. Matúš Harvan Xen Live Migration 1

Xen Live Migration. Networks and Distributed Systems Seminar, 24 April 2006. Matúš Harvan Xen Live Migration 1 Xen Live Migration Matúš Harvan Networks and Distributed Systems Seminar, 24 April 2006 Matúš Harvan Xen Live Migration 1 Outline 1 Xen Overview 2 Live migration General Memory, Network, Storage Migration

More information

Experimental Investigation Decentralized IaaS Cloud Architecture Open Stack with CDT

Experimental Investigation Decentralized IaaS Cloud Architecture Open Stack with CDT Experimental Investigation Decentralized IaaS Cloud Architecture Open Stack with CDT S. Gobinath, S. Saravanan PG Scholar, CSE Dept, M.Kumarasamy College of Engineering, Karur, India 1 Assistant Professor,

More information

EWeb: Highly Scalable Client Transparent Fault Tolerant System for Cloud based Web Applications

EWeb: Highly Scalable Client Transparent Fault Tolerant System for Cloud based Web Applications ECE6102 Dependable Distribute Systems, Fall2010 EWeb: Highly Scalable Client Transparent Fault Tolerant System for Cloud based Web Applications Deepal Jayasinghe, Hyojun Kim, Mohammad M. Hossain, Ali Payani

More information

International Journal of Computer Science Trends and Technology (IJCST) Volume 3 Issue 3, May-June 2015

International Journal of Computer Science Trends and Technology (IJCST) Volume 3 Issue 3, May-June 2015 RESEARCH ARTICLE OPEN ACCESS Ensuring Reliability and High Availability in Cloud by Employing a Fault Tolerance Enabled Load Balancing Algorithm G.Gayathri [1], N.Prabakaran [2] Department of Computer

More information

A Dynamic Resource Management with Energy Saving Mechanism for Supporting Cloud Computing

A Dynamic Resource Management with Energy Saving Mechanism for Supporting Cloud Computing A Dynamic Resource Management with Energy Saving Mechanism for Supporting Cloud Computing Liang-Teh Lee, Kang-Yuan Liu, Hui-Yang Huang and Chia-Ying Tseng Department of Computer Science and Engineering,

More information

Migration of Virtual Machines for Better Performance in Cloud Computing Environment

Migration of Virtual Machines for Better Performance in Cloud Computing Environment Migration of Virtual Machines for Better Performance in Cloud Computing Environment J.Sreekanth 1, B.Santhosh Kumar 2 PG Scholar, Dept. of CSE, G Pulla Reddy Engineering College, Kurnool, Andhra Pradesh,

More information

EC2 Performance Analysis for Resource Provisioning of Service-Oriented Applications

EC2 Performance Analysis for Resource Provisioning of Service-Oriented Applications EC2 Performance Analysis for Resource Provisioning of Service-Oriented Applications Jiang Dejun 1,2 Guillaume Pierre 1 Chi-Hung Chi 2 1 VU University Amsterdam 2 Tsinghua University Beijing Abstract. Cloud

More information

Performance Isolation of a Misbehaving Virtual Machine with Xen, VMware and Solaris Containers

Performance Isolation of a Misbehaving Virtual Machine with Xen, VMware and Solaris Containers Performance Isolation of a Misbehaving Virtual Machine with Xen, VMware and Solaris Containers Todd Deshane, Demetrios Dimatos, Gary Hamilton, Madhujith Hapuarachchi, Wenjin Hu, Michael McCabe, Jeanna

More information

Efficient DNS based Load Balancing for Bursty Web Application Traffic

Efficient DNS based Load Balancing for Bursty Web Application Traffic ISSN Volume 1, No.1, September October 2012 International Journal of Science the and Internet. Applied However, Information this trend leads Technology to sudden burst of Available Online at http://warse.org/pdfs/ijmcis01112012.pdf

More information

Collaborative & Integrated Network & Systems Management: Management Using Grid Technologies

Collaborative & Integrated Network & Systems Management: Management Using Grid Technologies 2011 International Conference on Computer Communication and Management Proc.of CSIT vol.5 (2011) (2011) IACSIT Press, Singapore Collaborative & Integrated Network & Systems Management: Management Using

More information

From Grid Computing to Cloud Computing & Security Issues in Cloud Computing

From Grid Computing to Cloud Computing & Security Issues in Cloud Computing From Grid Computing to Cloud Computing & Security Issues in Cloud Computing Rajendra Kumar Dwivedi Assistant Professor (Department of CSE), M.M.M. Engineering College, Gorakhpur (UP), India E-mail: rajendra_bhilai@yahoo.com

More information

Fault-Tolerant Framework for Load Balancing System

Fault-Tolerant Framework for Load Balancing System Fault-Tolerant Framework for Load Balancing System Y. K. LIU, L.M. CHENG, L.L.CHENG Department of Electronic Engineering City University of Hong Kong Tat Chee Avenue, Kowloon, Hong Kong SAR HONG KONG Abstract:

More information

High Performance Cluster Support for NLB on Window

High Performance Cluster Support for NLB on Window High Performance Cluster Support for NLB on Window [1]Arvind Rathi, [2] Kirti, [3] Neelam [1]M.Tech Student, Department of CSE, GITM, Gurgaon Haryana (India) arvindrathi88@gmail.com [2]Asst. Professor,

More information

Dynamic resource management for energy saving in the cloud computing environment

Dynamic resource management for energy saving in the cloud computing environment Dynamic resource management for energy saving in the cloud computing environment Liang-Teh Lee, Kang-Yuan Liu, and Hui-Yang Huang Department of Computer Science and Engineering, Tatung University, Taiwan

More information

IaaS Cloud Architectures: Virtualized Data Centers to Federated Cloud Infrastructures

IaaS Cloud Architectures: Virtualized Data Centers to Federated Cloud Infrastructures IaaS Cloud Architectures: Virtualized Data Centers to Federated Cloud Infrastructures Dr. Sanjay P. Ahuja, Ph.D. 2010-14 FIS Distinguished Professor of Computer Science School of Computing, UNF Introduction

More information

Best Practices for Managing Virtualized Environments

Best Practices for Managing Virtualized Environments WHITE PAPER Introduction... 2 Reduce Tool and Process Sprawl... 2 Control Virtual Server Sprawl... 3 Effectively Manage Network Stress... 4 Reliably Deliver Application Services... 5 Comprehensively Manage

More information

Dynamic Load Balancing of Virtual Machines using QEMU-KVM

Dynamic Load Balancing of Virtual Machines using QEMU-KVM Dynamic Load Balancing of Virtual Machines using QEMU-KVM Akshay Chandak Krishnakant Jaju Technology, College of Engineering, Pune. Maharashtra, India. Akshay Kanfade Pushkar Lohiya Technology, College

More information

Automation, Manageability, Architecture, Virtualization, data center, virtual machine, placement

Automation, Manageability, Architecture, Virtualization, data center, virtual machine, placement Autonomic Virtual Machine Placement in the Data Center Chris Hyser, Bret McKee, Rob Gardner, Brian J. Watson HP Laboratories HPL-2007-189 February 26, 2008* Automation, Manageability, Architecture, Virtualization,

More information

Load Balancing on a Non-dedicated Heterogeneous Network of Workstations

Load Balancing on a Non-dedicated Heterogeneous Network of Workstations Load Balancing on a Non-dedicated Heterogeneous Network of Workstations Dr. Maurice Eggen Nathan Franklin Department of Computer Science Trinity University San Antonio, Texas 78212 Dr. Roger Eggen Department

More information

USING VIRTUAL MACHINE REPLICATION FOR DYNAMIC CONFIGURATION OF MULTI-TIER INTERNET SERVICES

USING VIRTUAL MACHINE REPLICATION FOR DYNAMIC CONFIGURATION OF MULTI-TIER INTERNET SERVICES USING VIRTUAL MACHINE REPLICATION FOR DYNAMIC CONFIGURATION OF MULTI-TIER INTERNET SERVICES Carlos Oliveira, Vinicius Petrucci, Orlando Loques Universidade Federal Fluminense Niterói, Brazil ABSTRACT In

More information

Balancing Server in Public Cloud Using AJAS Algorithm

Balancing Server in Public Cloud Using AJAS Algorithm Balancing Server in Public Cloud Using AJAS Algorithm Ramya.B 1, Pragaladan R 2, M.Phil Part-Time Research Scholar, Assistant Professor Department of Computer Science, Department of Computer Science, Sri

More information

Detecting Computer Worms in the Cloud

Detecting Computer Worms in the Cloud Detecting Computer Worms in the Cloud Sebastian Biedermann and Stefan Katzenbeisser Security Engineering Group Department of Computer Science Technische Universität Darmstadt {biedermann,katzenbeisser}@seceng.informatik.tu-darmstadt.de

More information

Allocation of Resources Dynamically in Data Centre for Cloud Environment

Allocation of Resources Dynamically in Data Centre for Cloud Environment Allocation of Resources Dynamically in Data Centre for Cloud Environment Mr.Pramod 1, Mr. Kumar Swamy 2, Mr. Sunitha B. S 3 ¹Computer Science & Engineering, EPCET, VTU, INDIA ² Computer Science & Engineering,

More information

Recommendations for Performance Benchmarking

Recommendations for Performance Benchmarking Recommendations for Performance Benchmarking Shikhar Puri Abstract Performance benchmarking of applications is increasingly becoming essential before deployment. This paper covers recommendations and best

More information

Models For Modeling and Measuring the Performance of a Xen Virtual Server

Models For Modeling and Measuring the Performance of a Xen Virtual Server Measuring and Modeling the Performance of the Xen VMM Jie Lu, Lev Makhlis, Jianjiun Chen BMC Software Inc. Waltham, MA 2451 Server virtualization technology provides an alternative for server consolidation

More information

Group Based Load Balancing Algorithm in Cloud Computing Virtualization

Group Based Load Balancing Algorithm in Cloud Computing Virtualization Group Based Load Balancing Algorithm in Cloud Computing Virtualization Rishi Bhardwaj, 2 Sangeeta Mittal, Student, 2 Assistant Professor, Department of Computer Science, Jaypee Institute of Information

More information

VMware vsphere: Install, Configure, Manage [V5.0]

VMware vsphere: Install, Configure, Manage [V5.0] VMware vsphere: Install, Configure, Manage [V5.0] Gain hands-on experience using VMware ESXi 5.0 and vcenter Server 5.0. In this hands-on, VMware -authorized course based on ESXi 5.0 and vcenter Server

More information

Optimizing Shared Resource Contention in HPC Clusters

Optimizing Shared Resource Contention in HPC Clusters Optimizing Shared Resource Contention in HPC Clusters Sergey Blagodurov Simon Fraser University Alexandra Fedorova Simon Fraser University Abstract Contention for shared resources in HPC clusters occurs

More information

Performance Comparison of Assignment Policies on Cluster-based E-Commerce Servers

Performance Comparison of Assignment Policies on Cluster-based E-Commerce Servers Performance Comparison of Assignment Policies on Cluster-based E-Commerce Servers Victoria Ungureanu Department of MSIS Rutgers University, 180 University Ave. Newark, NJ 07102 USA Benjamin Melamed Department

More information

Multilevel Load Balancing in NUMA Computers

Multilevel Load Balancing in NUMA Computers FACULDADE DE INFORMÁTICA PUCRS - Brazil http://www.pucrs.br/inf/pos/ Multilevel Load Balancing in NUMA Computers M. Corrêa, R. Chanin, A. Sales, R. Scheer, A. Zorzo Technical Report Series Number 049 July,

More information

Chapter 1 - Web Server Management and Cluster Topology

Chapter 1 - Web Server Management and Cluster Topology Objectives At the end of this chapter, participants will be able to understand: Web server management options provided by Network Deployment Clustered Application Servers Cluster creation and management

More information

A Novel Method for Resource Allocation in Cloud Computing Using Virtual Machines

A Novel Method for Resource Allocation in Cloud Computing Using Virtual Machines A Novel Method for Resource Allocation in Cloud Computing Using Virtual Machines Ch.Anusha M.Tech, Dr.K.Babu Rao, M.Tech, Ph.D Professor, MR. M.Srikanth Asst Professor & HOD, Abstract: Cloud computing

More information

How To Balance A Web Server With Remaining Capacity

How To Balance A Web Server With Remaining Capacity Remaining Capacity Based Load Balancing Architecture for Heterogeneous Web Server System Tsang-Long Pao Dept. Computer Science and Engineering Tatung University Taipei, ROC Jian-Bo Chen Dept. Computer

More information

An Evaluation of Economy-based Resource Trading and Scheduling on Computational Power Grids for Parameter Sweep Applications

An Evaluation of Economy-based Resource Trading and Scheduling on Computational Power Grids for Parameter Sweep Applications An Evaluation of Economy-based Resource Trading and Scheduling on Computational Power Grids for Parameter Sweep Applications Rajkumar Buyya, Jonathan Giddy, and David Abramson School of Computer Science

More information

Enabling Technologies for Distributed Computing

Enabling Technologies for Distributed Computing Enabling Technologies for Distributed Computing Dr. Sanjay P. Ahuja, Ph.D. Fidelity National Financial Distinguished Professor of CIS School of Computing, UNF Multi-core CPUs and Multithreading Technologies

More information

Running VirtualCenter in a Virtual Machine

Running VirtualCenter in a Virtual Machine VMWARE TECHNICAL NOTE VirtualCenter 2.x Running VirtualCenter in a Virtual Machine Running VirtualCenter in a virtual machine is fully supported by VMware to the same degree as if it were installed on

More information

CLOUD COMPUTING. DAV University, Jalandhar, Punjab, India. DAV University, Jalandhar, Punjab, India

CLOUD COMPUTING. DAV University, Jalandhar, Punjab, India. DAV University, Jalandhar, Punjab, India CLOUD COMPUTING 1 Er. Simar Preet Singh, 2 Er. Anshu Joshi 1 Assistant Professor, Computer Science & Engineering, DAV University, Jalandhar, Punjab, India 2 Research Scholar, Computer Science & Engineering,

More information

What s New in this Brief

What s New in this Brief B r i e f Licensing Microsoft Windows Server 2008 to Run with Virtualization Technologies VMware s ESX Server, VMware s VMotion, Microsoft s System Center Virtualization Machine Manager, and Parallels

More information

REM-Rocks: A Runtime Environment Migration Scheme for Rocks based Linux HPC Clusters

REM-Rocks: A Runtime Environment Migration Scheme for Rocks based Linux HPC Clusters REM-Rocks: A Runtime Environment Migration Scheme for Rocks based Linux HPC Clusters Tong Liu, Saeed Iqbal, Yung-Chin Fang, Onur Celebioglu, Victor Masheyakhi and Reza Rooholamini Dell Inc. {Tong_Liu,

More information

Flauncher and DVMS Deploying and Scheduling Thousands of Virtual Machines on Hundreds of Nodes Distributed Geographically

Flauncher and DVMS Deploying and Scheduling Thousands of Virtual Machines on Hundreds of Nodes Distributed Geographically Flauncher and Deploying and Scheduling Thousands of Virtual Machines on Hundreds of Nodes Distributed Geographically Daniel Balouek, Adrien Lèbre, Flavien Quesnel To cite this version: Daniel Balouek,

More information

Distributed resource discovery and management in the XenoServers platform

Distributed resource discovery and management in the XenoServers platform Distributed resource discovery and management in the XenoServers platform Evangelos Kotsovinos and Timothy L Harris University of Cambridge Computer Laboratory Contact: E Kotsovinos, Computer Laboratory

More information

Distributed Systems and Recent Innovations: Challenges and Benefits

Distributed Systems and Recent Innovations: Challenges and Benefits Distributed Systems and Recent Innovations: Challenges and Benefits 1. Introduction Krishna Nadiminti, Marcos Dias de Assunção, and Rajkumar Buyya Grid Computing and Distributed Systems Laboratory Department

More information

Scientific Cloud Computing: Early Definition and Experience

Scientific Cloud Computing: Early Definition and Experience The 10th IEEE International Conference on High Performance Computing and Communications Scientific Cloud Computing: Early Definition and Experience Lizhe Wang, Jie Tao, Marcel Kunze Institute for Scientific

More information

Virtualization for Future Internet

Virtualization for Future Internet Virtualization for Future Internet 2010.02.23 Korea University Chuck Yoo (hxy@os.korea.ac.kr) Why Virtualization Internet today Pro and con Your wonderful research results Mostly with simulation Deployment

More information

The EMSX Platform. A Modular, Scalable, Efficient, Adaptable Platform to Manage Multi-technology Networks. A White Paper.

The EMSX Platform. A Modular, Scalable, Efficient, Adaptable Platform to Manage Multi-technology Networks. A White Paper. The EMSX Platform A Modular, Scalable, Efficient, Adaptable Platform to Manage Multi-technology Networks A White Paper November 2002 Abstract: The EMSX Platform is a set of components that together provide

More information

PERFORMANCE ANALYSIS OF PaaS CLOUD COMPUTING SYSTEM

PERFORMANCE ANALYSIS OF PaaS CLOUD COMPUTING SYSTEM PERFORMANCE ANALYSIS OF PaaS CLOUD COMPUTING SYSTEM Akmal Basha 1 Krishna Sagar 2 1 PG Student,Department of Computer Science and Engineering, Madanapalle Institute of Technology & Science, India. 2 Associate

More information

A Comparison on Current Distributed File Systems for Beowulf Clusters

A Comparison on Current Distributed File Systems for Beowulf Clusters A Comparison on Current Distributed File Systems for Beowulf Clusters Rafael Bohrer Ávila 1 Philippe Olivier Alexandre Navaux 2 Yves Denneulin 3 Abstract This paper presents a comparison on current file

More information

Elastic Load Balancing in Cloud Storage

Elastic Load Balancing in Cloud Storage Elastic Load Balancing in Cloud Storage Surabhi Jain, Deepak Sharma (Lecturer, Department of Computer Science, Lovely Professional University, Phagwara-144402) (Assistant Professor, Department of Computer

More information

Resource Scalability for Efficient Parallel Processing in Cloud

Resource Scalability for Efficient Parallel Processing in Cloud Resource Scalability for Efficient Parallel Processing in Cloud ABSTRACT Govinda.K #1, Abirami.M #2, Divya Mercy Silva.J #3 #1 SCSE, VIT University #2 SITE, VIT University #3 SITE, VIT University In the

More information

Cluster, Grid, Cloud Concepts

Cluster, Grid, Cloud Concepts Cluster, Grid, Cloud Concepts Kalaiselvan.K Contents Section 1: Cluster Section 2: Grid Section 3: Cloud Cluster An Overview Need for a Cluster Cluster categorizations A computer cluster is a group of

More information

LSKA 2010 Survey Report Job Scheduler

LSKA 2010 Survey Report Job Scheduler LSKA 2010 Survey Report Job Scheduler Graduate Institute of Communication Engineering {r98942067, r98942112}@ntu.edu.tw March 31, 2010 1. Motivation Recently, the computing becomes much more complex. However,

More information

From Grid Computing to Cloud Computing & Security Issues in Cloud Computing

From Grid Computing to Cloud Computing & Security Issues in Cloud Computing From Grid Computing to Cloud Computing & Security Issues in Cloud Computing Rajendra Kumar Dwivedi Department of CSE, M.M.M. Engineering College, Gorakhpur (UP), India 273010 rajendra_bhilai@yahoo.com

More information

A New Scalable Server Construction Method Based On. Virtual Machine Transfer and Duplication

A New Scalable Server Construction Method Based On. Virtual Machine Transfer and Duplication A New Scalable Server Construction Method Based On Virtual Machine Transfer and Duplication Yuko KAMIYA and Toshihiko SHIMOKAWA Graduate School of Information Science, Kyushu Sangyo University, JAPAN Faculty

More information

The Importance of Software License Server Monitoring

The Importance of Software License Server Monitoring The Importance of Software License Server Monitoring NetworkComputer How Shorter Running Jobs Can Help In Optimizing Your Resource Utilization White Paper Introduction Semiconductor companies typically

More information

WHITE PAPER The Storage Holy Grail: Decoupling Performance from Capacity

WHITE PAPER The Storage Holy Grail: Decoupling Performance from Capacity WHITE PAPER The Storage Holy Grail: Decoupling Performance from Capacity Technical White Paper 1 The Role of a Flash Hypervisor in Today s Virtual Data Center Virtualization has been the biggest trend

More information

Infrastructure as a Service (IaaS)

Infrastructure as a Service (IaaS) Infrastructure as a Service (IaaS) (ENCS 691K Chapter 4) Roch Glitho, PhD Associate Professor and Canada Research Chair My URL - http://users.encs.concordia.ca/~glitho/ References 1. R. Moreno et al.,

More information

A Middleware Strategy to Survive Compute Peak Loads in Cloud

A Middleware Strategy to Survive Compute Peak Loads in Cloud A Middleware Strategy to Survive Compute Peak Loads in Cloud Sasko Ristov Ss. Cyril and Methodius University Faculty of Information Sciences and Computer Engineering Skopje, Macedonia Email: sashko.ristov@finki.ukim.mk

More information

A Distributed Render Farm System for Animation Production

A Distributed Render Farm System for Animation Production A Distributed Render Farm System for Animation Production Jiali Yao, Zhigeng Pan *, Hongxin Zhang State Key Lab of CAD&CG, Zhejiang University, Hangzhou, 310058, China {yaojiali, zgpan, zhx}@cad.zju.edu.cn

More information

Is Virtualization Killing SSI Research?

Is Virtualization Killing SSI Research? Is Virtualization Killing SSI Research? Jérôme Gallard Paris Project-Team Dinard November 2007 Supervisor : Christine Morin Co-supervisor: Adrien Lèbre My subject! ;) Reliability and performance of execution

More information

Index Terms : Load rebalance, distributed file systems, clouds, movement cost, load imbalance, chunk.

Index Terms : Load rebalance, distributed file systems, clouds, movement cost, load imbalance, chunk. Load Rebalancing for Distributed File Systems in Clouds. Smita Salunkhe, S. S. Sannakki Department of Computer Science and Engineering KLS Gogte Institute of Technology, Belgaum, Karnataka, India Affiliated

More information

Energetic Resource Allocation Framework Using Virtualization in Cloud

Energetic Resource Allocation Framework Using Virtualization in Cloud Energetic Resource Allocation Framework Using Virtualization in Ms.K.Guna *1, Ms.P.Saranya M.E *2 1 (II M.E(CSE)) Student Department of Computer Science and Engineering, 2 Assistant Professor Department

More information

MEASURING WORKLOAD PERFORMANCE IS THE INFRASTRUCTURE A PROBLEM?

MEASURING WORKLOAD PERFORMANCE IS THE INFRASTRUCTURE A PROBLEM? MEASURING WORKLOAD PERFORMANCE IS THE INFRASTRUCTURE A PROBLEM? Ashutosh Shinde Performance Architect ashutosh_shinde@hotmail.com Validating if the workload generated by the load generating tools is applied

More information

JOB ORIENTED VMWARE TRAINING INSTITUTE IN CHENNAI

JOB ORIENTED VMWARE TRAINING INSTITUTE IN CHENNAI JOB ORIENTED VMWARE TRAINING INSTITUTE IN CHENNAI Job oriented VMWARE training is offered by Peridot Systems in Chennai. Training in our institute gives you strong foundation on cloud computing by incrementing

More information

Design Notes for an Efficient Password-Authenticated Key Exchange Implementation Using Human-Memorable Passwords

Design Notes for an Efficient Password-Authenticated Key Exchange Implementation Using Human-Memorable Passwords Design Notes for an Efficient Password-Authenticated Key Exchange Implementation Using Human-Memorable Passwords Author: Paul Seymer CMSC498a Contents 1 Background... 2 1.1 HTTP 1.0/1.1... 2 1.2 Password

More information

Virtual Machine Based Resource Allocation For Cloud Computing Environment

Virtual Machine Based Resource Allocation For Cloud Computing Environment Virtual Machine Based Resource Allocation For Cloud Computing Environment D.Udaya Sree M.Tech (CSE) Department Of CSE SVCET,Chittoor. Andra Pradesh, India Dr.J.Janet Head of Department Department of CSE

More information

Enabling Technologies for Distributed and Cloud Computing

Enabling Technologies for Distributed and Cloud Computing Enabling Technologies for Distributed and Cloud Computing Dr. Sanjay P. Ahuja, Ph.D. 2010-14 FIS Distinguished Professor of Computer Science School of Computing, UNF Multi-core CPUs and Multithreading

More information

CHAPTER 2 MODELLING FOR DISTRIBUTED NETWORK SYSTEMS: THE CLIENT- SERVER MODEL

CHAPTER 2 MODELLING FOR DISTRIBUTED NETWORK SYSTEMS: THE CLIENT- SERVER MODEL CHAPTER 2 MODELLING FOR DISTRIBUTED NETWORK SYSTEMS: THE CLIENT- SERVER MODEL This chapter is to introduce the client-server model and its role in the development of distributed network systems. The chapter

More information

Grid Computing Vs. Cloud Computing

Grid Computing Vs. Cloud Computing International Journal of Information and Computation Technology. ISSN 0974-2239 Volume 3, Number 6 (2013), pp. 577-582 International Research Publications House http://www. irphouse.com /ijict.htm Grid

More information

A Review of Load Balancing Algorithms for Cloud Computing

A Review of Load Balancing Algorithms for Cloud Computing www.ijecs.in International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume - 3 Issue -9 September, 2014 Page No. 8297-8302 A Review of Load Balancing Algorithms for Cloud Computing Dr.G.N.K.Sureshbabu

More information

Copyright www.agileload.com 1

Copyright www.agileload.com 1 Copyright www.agileload.com 1 INTRODUCTION Performance testing is a complex activity where dozens of factors contribute to its success and effective usage of all those factors is necessary to get the accurate

More information

There are a number of factors that increase the risk of performance problems in complex computer and software systems, such as e-commerce systems.

There are a number of factors that increase the risk of performance problems in complex computer and software systems, such as e-commerce systems. ASSURING PERFORMANCE IN E-COMMERCE SYSTEMS Dr. John Murphy Abstract Performance Assurance is a methodology that, when applied during the design and development cycle, will greatly increase the chances

More information

Ensuring Security in Cloud with Multi-Level IDS and Log Management System

Ensuring Security in Cloud with Multi-Level IDS and Log Management System Ensuring Security in Cloud with Multi-Level IDS and Log Management System 1 Prema Jain, 2 Ashwin Kumar PG Scholar, Mangalore Institute of Technology & Engineering, Moodbidri, Karnataka1, Assistant Professor,

More information

Cloud Computing: a Perspective Study

Cloud Computing: a Perspective Study Cloud Computing: a Perspective Study Lizhe WANG, Gregor von LASZEWSKI, Younge ANDREW, Xi HE Service Oriented Cyberinfrastruture Lab, Rochester Inst. of Tech. Lomb Memorial Drive, Rochester, NY 14623, U.S.

More information

Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at

Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at distributing load b. QUESTION: What is the context? i. How

More information

Dynamic Creation and Placement of Virtual Machine Using CloudSim

Dynamic Creation and Placement of Virtual Machine Using CloudSim Dynamic Creation and Placement of Virtual Machine Using CloudSim Vikash Rao Pahalad Singh College of Engineering, Balana, India Abstract --Cloud Computing becomes a new trend in computing. The IaaS(Infrastructure

More information

Web Service Based Data Management for Grid Applications

Web Service Based Data Management for Grid Applications Web Service Based Data Management for Grid Applications T. Boehm Zuse-Institute Berlin (ZIB), Berlin, Germany Abstract Web Services play an important role in providing an interface between end user applications

More information

Datasheet iscsi Protocol

Datasheet iscsi Protocol Protocol with DCB PROTOCOL PACKAGE Industry s premiere validation system for SAN technologies Overview Load DynamiX offers SCSI over TCP/IP transport () support to its existing powerful suite of file,

More information

Virtual machine interface. Operating system. Physical machine interface

Virtual machine interface. Operating system. Physical machine interface Software Concepts User applications Operating system Hardware Virtual machine interface Physical machine interface Operating system: Interface between users and hardware Implements a virtual machine that

More information

Enhancing the Performance of Live Migration of Virtual Machine s with WSClock Replacement Algorithm

Enhancing the Performance of Live Migration of Virtual Machine s with WSClock Replacement Algorithm Enhancing the Performance of Live Migration of Virtual Machine s with WSClock Replacement Algorithm C.Sagana M.Geetha Dr R.C.Suganthe PG student, Assistant Professor, Professor, Dept of CSE, Dept of CSE

More information

INCREASING SERVER UTILIZATION AND ACHIEVING GREEN COMPUTING IN CLOUD

INCREASING SERVER UTILIZATION AND ACHIEVING GREEN COMPUTING IN CLOUD INCREASING SERVER UTILIZATION AND ACHIEVING GREEN COMPUTING IN CLOUD M.Rajeswari 1, M.Savuri Raja 2, M.Suganthy 3 1 Master of Technology, Department of Computer Science & Engineering, Dr. S.J.S Paul Memorial

More information

Cloud Storage. Parallels. Performance Benchmark Results. White Paper. www.parallels.com

Cloud Storage. Parallels. Performance Benchmark Results. White Paper. www.parallels.com Parallels Cloud Storage White Paper Performance Benchmark Results www.parallels.com Table of Contents Executive Summary... 3 Architecture Overview... 3 Key Features... 4 No Special Hardware Requirements...

More information

Xen and the Art of Distributed Virtual Machine Management

Xen and the Art of Distributed Virtual Machine Management QuickTime and a TIFF (Uncompressed) decompressor are needed to see this picture. Xen and the Art of Distributed Virtual Machine Management Dr. Greg Lavender Mr. Adam Zacharski Department of Computer Sciences

More information

Distributed and Cloud Computing

Distributed and Cloud Computing Distributed and Cloud Computing K. Hwang, G. Fox and J. Dongarra Chapter 3: Virtual Machines and Virtualization of Clusters and datacenters Adapted from Kai Hwang University of Southern California March

More information

OpenMosix Presented by Dr. Moshe Bar and MAASK [01]

OpenMosix Presented by Dr. Moshe Bar and MAASK [01] OpenMosix Presented by Dr. Moshe Bar and MAASK [01] openmosix is a kernel extension for single-system image clustering. openmosix [24] is a tool for a Unix-like kernel, such as Linux, consisting of adaptive

More information

Implementing Security on virtualized network storage environment

Implementing Security on virtualized network storage environment International Journal of Education and Research Vol. 2 No. 4 April 2014 Implementing Security on virtualized network storage environment Benard O. Osero, David G. Mwathi Chuka University bosero@chuka.ac.ke

More information

MEASURING PERFORMANCE OF DYNAMIC LOAD BALANCING ALGORITHMS IN DISTRIBUTED COMPUTING APPLICATIONS

MEASURING PERFORMANCE OF DYNAMIC LOAD BALANCING ALGORITHMS IN DISTRIBUTED COMPUTING APPLICATIONS MEASURING PERFORMANCE OF DYNAMIC LOAD BALANCING ALGORITHMS IN DISTRIBUTED COMPUTING APPLICATIONS Priyesh Kanungo 1 Professor and Senior Systems Engineer (Computer Centre), School of Computer Science and

More information

Veeam Cloud Connect. Version 8.0. Administrator Guide

Veeam Cloud Connect. Version 8.0. Administrator Guide Veeam Cloud Connect Version 8.0 Administrator Guide April, 2015 2015 Veeam Software. All rights reserved. All trademarks are the property of their respective owners. No part of this publication may be

More information

Building Scalable Applications Using Microsoft Technologies

Building Scalable Applications Using Microsoft Technologies Building Scalable Applications Using Microsoft Technologies Padma Krishnan Senior Manager Introduction CIOs lay great emphasis on application scalability and performance and rightly so. As business grows,

More information

Efficient and Enhanced Load Balancing Algorithms in Cloud Computing

Efficient and Enhanced Load Balancing Algorithms in Cloud Computing , pp.9-14 http://dx.doi.org/10.14257/ijgdc.2015.8.2.02 Efficient and Enhanced Load Balancing Algorithms in Cloud Computing Prabhjot Kaur and Dr. Pankaj Deep Kaur M. Tech, CSE P.H.D prabhjotbhullar22@gmail.com,

More information

A Taxonomy and Survey of Energy-Efficient Data Centers and Cloud Computing Systems

A Taxonomy and Survey of Energy-Efficient Data Centers and Cloud Computing Systems A Taxonomy and Survey of Energy-Efficient Data Centers and Cloud Computing Systems Anton Beloglazov, Rajkumar Buyya, Young Choon Lee, and Albert Zomaya Present by Leping Wang 1/25/2012 Outline Background

More information

Cloud computing: A Perspective study

Cloud computing: A Perspective study Rochester Institute of Technology RIT Scholar Works Presentations and other scholarship 2008 Cloud computing: A Perspective study Lizhe Wang Gregor von Laszewski Marcel Kunze Jie Tao Follow this and additional

More information

Cloud Computing for Control Systems CERN Openlab Summer Student Program 9/9/2011 ARSALAAN AHMED SHAIKH

Cloud Computing for Control Systems CERN Openlab Summer Student Program 9/9/2011 ARSALAAN AHMED SHAIKH Cloud Computing for Control Systems CERN Openlab Summer Student Program 9/9/2011 ARSALAAN AHMED SHAIKH CONTENTS Introduction... 4 System Components... 4 OpenNebula Cloud Management Toolkit... 4 VMware

More information

Payment minimization and Error-tolerant Resource Allocation for Cloud System Using equally spread current execution load

Payment minimization and Error-tolerant Resource Allocation for Cloud System Using equally spread current execution load Payment minimization and Error-tolerant Resource Allocation for Cloud System Using equally spread current execution load Pooja.B. Jewargi Prof. Jyoti.Patil Department of computer science and engineering,

More information

Principles and characteristics of distributed systems and environments

Principles and characteristics of distributed systems and environments Principles and characteristics of distributed systems and environments Definition of a distributed system Distributed system is a collection of independent computers that appears to its users as a single

More information

Content-Aware Load Balancing using Direct Routing for VOD Streaming Service

Content-Aware Load Balancing using Direct Routing for VOD Streaming Service Content-Aware Load Balancing using Direct Routing for VOD Streaming Service Young-Hwan Woo, Jin-Wook Chung, Seok-soo Kim Dept. of Computer & Information System, Geo-chang Provincial College, Korea School

More information

BlackBerry Enterprise Service 10. Secure Work Space for ios and Android Version: 10.1.1. Security Note

BlackBerry Enterprise Service 10. Secure Work Space for ios and Android Version: 10.1.1. Security Note BlackBerry Enterprise Service 10 Secure Work Space for ios and Android Version: 10.1.1 Security Note Published: 2013-06-21 SWD-20130621110651069 Contents 1 About this guide...4 2 What is BlackBerry Enterprise

More information

IDENTIFYING AND OPTIMIZING DATA DUPLICATION BY EFFICIENT MEMORY ALLOCATION IN REPOSITORY BY SINGLE INSTANCE STORAGE

IDENTIFYING AND OPTIMIZING DATA DUPLICATION BY EFFICIENT MEMORY ALLOCATION IN REPOSITORY BY SINGLE INSTANCE STORAGE IDENTIFYING AND OPTIMIZING DATA DUPLICATION BY EFFICIENT MEMORY ALLOCATION IN REPOSITORY BY SINGLE INSTANCE STORAGE 1 M.PRADEEP RAJA, 2 R.C SANTHOSH KUMAR, 3 P.KIRUTHIGA, 4 V. LOGESHWARI 1,2,3 Student,

More information

Performance White Paper

Performance White Paper Sitecore Experience Platform 8.1 Performance White Paper Rev: March 11, 2016 Sitecore Experience Platform 8.1 Performance White Paper Sitecore Experience Platform 8.1 Table of contents Table of contents...

More information