Delivering SDS simplicity and extreme performance Real-World SDS implementation of getting most out of limited hardware Murat Karslioglu Director Storage Systems Nexenta Systems October 2013 1
Agenda Key Takeaways Introduction Performance Results Conclusion Q & A October 2013 2
Key Takeaways VDI as a case study of SDS delivering multitenancy and on-demand provisioning Remove storage from the VDI admin's plate Get higher VDI density and better performance out of the limited hardware resources October 2013 3
Consolidate. Simplify. Virtualize. Monitor We picked an affordable branch office server: Limited resources, NOT a great fit for VDI Intel Xeon E5-2400 series 6 core processor 48 Gigabytes of RAM Three 2.5 size HDDs (No SSDs) October 2013 4
Challanges HIGH STORAGE COST VDI (storage) PERFORMANCE IS BAD TOO COMPLEX BAD END USER EXPERIENCE LIMITED RESOURCES FAILED POCs October 2013 5
The storage guessing game Connection Broker Connection Agent Connection Agent Connection Agent Connection Agent Connection Agent Management Server Hypervisor Physical Servers Shared Storage October 2013 6
How does NV4V Remove the storage guessing game? Integrate VDI and Storage In depth integration between NexentaVSA and VMware Horizon View, vsphere, vcenter New features to optimize storage A user friendly application to simplify and automate NAS VAAI Integration Real-world concrete SDS implementation October 2013 7
NV4V as software-defined-storage Deploy Measure Configure Step function increments to meet performance requirements (Bandwidth, latency and IOPS) October 2013 8
NV4V High-level Architecture October 2013 9
NV4V VDI Deployment Overview 1. NV4V->vCenter: Provision NFS network (N/A with external NexentaStor) 2. NV4V->vCenter: Provision VSA. (N/A with external NexentaStor) 3. NV4V->vCenter: Create and attach VMDK datastores, Power on VSA. (N/A with external NexentaStor) 4. NV4V->VSA: Create zpools and NFS shares. (Opt. with external NexentaStor) 5. NV4V->View: Deploy desktop pool. ESXi Cluster (NV4V Communicates with Desktop Nexenta Agents For Benchmark and Calibration) VMDKs on ESXi host disks October 2013 10
NV4V VDI Deployment Overview Process 1 Create VMDK(s) for VSA syspool (resilver if mirrored) 2 Create resource pool for VSA 3 Clone NexentaStor VSA template 4 Reconfigure VSA - Set resources, reservations and limits Point of View Event 5 Power on VSA 6 Confirm VMware Tools on VSA 7 Assign DHCP address to VSA Network interface 8 Verify DHCP and Reverse mapping Activity Event VSA Event Process 9 Create VMDK datastores for data 10 Attach datastores to the VSA, one by one 11 Create Port Group or use existing one for NFS network 12 Configure port group - set MTU to 9000 for NFS network 13 Configure ZFS tunables and reboot 14 Linked-clones only: Create and configure zpools, by default two zpools, Replica and Desktop; Only one for all SSD desktop pool Point of View ZpoolHistory, Event ZpoolHistory, Event 15 Share NFS filesystems ZpoolHistory, Event 16 Mount NFS datastores to all hypervisors Process 17 Start deployment through VMware View 18 Linked-clones only: Copy Replica image from snapshot to NFS Replica filesystem 19 Linked-clones only: Create VMs, store linked clones in NFS Desktop fielsystem 20 Full-clones only: Clone desktop image from template to NFS Desktop filesystem once for every desktop Point of View Activity Event Activity 21 Customize desktops Activity Event, VDI 22 Finish when target number of desktop pool size is hit 23 Entitlements Activity Event October 2013 11
NV4V VDI Deployment Overview VSA VMDK s and NFS Shares October 2013 12
Improving performance with NV4V Nexenta NV4V + Server + VMware View = Perfect Branch Office Solution 3x Higher Density Tested with LoginVSI 11x Better End-user Experience Tested with IoMeter (75%Write) With NV4V Local Disk Medium Workload 55 Desktops* 18 Desktops Heavy Workload 37 Desktops 12 Desktops With NV4V 55 Desktops With NV4V 18 Desktops Local Disk 18 Desktops** IoMeter Total IOPS 2343 IOPS 2160 IOPS 198 IOPS IoMeter IOPS/Desktop 42.6 IOPS 120 IOPS 11 IOPS Simplified deployment On-demand storage Monitoring Backup/Restore NAS VAAI Software RAID *VSImax not reached, 55 desktop is due to memory limitation on Cisco UCS E Series platform ** VSImax 18 with local disk Inline Compression Caching on memory (ARC) Other ZFS Benefits October 2013 13
Improving performance with NV4V Speed up full clone deployment First-time in world s history NV4V utilizes NAS VAAI to provide ZFS to CoW-clone files to deploy persistent VM images much faster, while at the same time saving on the storage capacity 5.4x faster to deploy full clones Comparison is for 24 full clone desktops 8.5x saving on storage capacity NAS VAAI utilizes enhanced Deduplication* w/ NAS VAAI w/o NAS VAAI Pure cloning time 2min 36sec 13min 36sec Clone&custimization 4min 38sec 18min 10sec w/ NAS VAAI w/o NAS VAAI Used Storage 48 GB 408 GB Dedup ratio x22.82 x1 Total deployment 2hours 38min 7hours 28min October 2013 14
Conclusion Desktop Pool running on local HDD take longer to login and start apps, causing increased CPU utilization, Single drive cannot handle more than 15 desktops efficiently. High random disk I/O causes CPU spikes resulting in dropped or frozen sessions, Storage is the most important component of virtualization, can also reduce CPU utilization, NV4V Benefit #1: SDS removes the storage guessing from admin s plate NV4V Benefit #2: Inline compression reduces writes up to 4x NV4V Benefit #3: Striping two drives doubles disk performance NV4V Benefit #4: NAS VAAI reduces full-clone deployment time and saves disk capacity NV4V Benefit #5: Reduced disk I/O and increased storage performance reduces CPU utilization NV4V Benefit #6: NV4V provides faster storage than world s fastest SSDs October 2013 15
LogiNVSI Medium Workload 55 linked-clone Desktops starting medium workload on local disk Before NV4V CPU Utilization 10 Desktops 66% 15 Desktops 81% 18 Desktops 88% 20 Desktops 90% 25 Desktops 100% >25 - Sessions dropped and desktops became unresponsive October 2013 16
LogiNVSI Medium Workload 55 linked-clone Desktops starting medium workload on NV4V With NV4V CPU Utilization 10 Desktops 45% 25 Desktops 75% 30 Desktops 78% 35 Desktops 80% 40 Desktops 81% 45 Desktops 82% 50 Desktops 84% 55 Desktops 88% October 2013 17
LogiNVSI Medium Workload 55 linked-clone Desktops running medium workload on NV4V With NV4V Recommended VSImax: VSImax not reached* Baseline = 2209 55 Desktops max 88% CPU utilization Desktops are highly responsive October 2013 18
LogiNVSI Heavy Workload 50 linked-clone Desktops starting heavy workload on NV4V With NV4V CPU Utilization 10 Desktops 47% 25 Desktops 76% 30 Desktops 79% 37 Desktops 85% 40 Desktops 90% 45 Desktops 92% 50 Desktops 94% October 2013 19
LogiNVSI Heavy Workload CPU utilization during 1 hour IOMETER test (/w NAS VAAI with full-clones) Threshold: < 90% Average utilization running IOmeter is ~84% October 2013 20
DISK BENCHMARK SDS solution turned slow HDDs into fastest SSD speed October 2013 21