Shared Distributed Memory : the Workspace Model

Transcription

1 Shared Distributed Memory : the Workspace Model Didier DONSEZ, Liming CHEN, Pascal FAUDEMAY Laboratoire HEUDIASYC - Université de Technologie de Compiègne Laboratoire MASI- Université Pierre et Marie Curie (Paris VI) {donsez,lchen}@hds.univ-compiegne.fr faudemay@masi.ibp.fr The OO-DBMS Architecture Application Clients Application Comm layer Comm layer Network Servers Comm layer Client Cnxn Comm layer Client Cnxn Database Archive Client / Server Model fits well to LANs with powerful Workstations ERSADS 95 Shared Distributed Memory : the Workspace Model 2

2 But there are some limitations... Does not fit to Heterogeneous LAN / WAN Networks Use of Check In / Check Out Mechanisms Need for Query Shipping Weak Clients Nomadic PCs, Walkstations Security and Confidentiality ERSADS 95 Shared Distributed Memory : the Workspace Model 3 The Nested Transaction Model [Moss85] T 1121 T 1121 T 111 T 112 T 111 T 112 T 221 T 222 T 11 T 12 T 11 T 12 T 21 T 22 T 1 T 1 T 2 T 221 T 222 T 21 T 22 DataBase T 2 ERSADS 95 Shared Distributed Memory : the Workspace Model 4

3 The Workspace Model Goal : Extend Moss Model Map Client and Server Processes on a graph of extended nested transactions Transaction Model the Workspace Model = + Process Model ERSADS 95 Shared Distributed Memory : the Workspace Model 5 WS Model : The Transaction Model simultaneous "sister" transactions Moss' Model + multiple parents transactions retaining or passing export T 8 T 8 T 8 access tov1 access to V2 T 6 T 7 T 6 T 7 T 6 T 7 T 3 T 4 T 3 T 4 T 5 = T 3 T 4 T 5 T 1 T 2 T 1 T 2 V 1 V 2 V 1 V 2 ERSADS 95 Shared Distributed Memory : the Workspace Model 6

4 WS Model : Retaining Transaction / Passing Transaction Retaining Export Moss' Commit Passing Export Behave as a Data Cache T commit T commit Passing Passing Retaining Passing Retaining ERSADS 95 Shared Distributed Memory : the Workspace Model 7 WS Model : The Process Model Restrict the graph of extended Transactions to efficiently implement the Workspace Process A Workspace Process to a to another N Sub Transactions Application 2 Server s Transaction Database View attached to 1 Main Transaction Volume Vj to Volume VO to Volume Vn a another ERSADS 95 Shared Distributed Memory : the Workspace Model 8

5 Instanciations of the Workspace Model A Workspace may be a standalone machine, a client, a server or both client and server Instanciations of some architectures : Standalone Database Client / Server Database Client / Server Database à la SHORE WAN Front-end ERSADS 95 Shared Distributed Memory : the Workspace Model 9 Client / Server Database Architecture the database is distributed on N servers : each client is directly connected to N servers a sequential application V1 Vn ERSADS 95 Shared Distributed Memory : the Workspace Model 10

6 Client / Server à la SHORE (U. of Wisconsin) each server and exports its part to the N-1 servers merges all parts and export them to its connected clients a sequential application V1 V n ERSADS 95 Shared Distributed Memory : the Workspace Model 11 Client / Server with WAN Front-End Client Workstation Client Workstation a sequential application LAN Front End WS Front End= Dept. Server Retaining Export for Check In / Check Out WAN Volume V0 Server Volume Vn Server V 0 Vn ERSADS 95 Shared Distributed Memory : the Workspace Model 12

7 Mixing Data and Query Shipping Data Shipping moves confidential data to unsecured clients Query Shipping involves bigger servers back to the Mainframe Model Merging Data-Query Shipping = 2 kinds of access permissions 1 for operation on server 1 for data export oid? lock/value query {oid nvalue} consistency data between Client and Server ERSADS 95 Shared Distributed Memory : the Workspace Model 13 Implementation 2 implementations WEA - YOODA True MultiThreading parallelism (CPU,IO) and asynchronous communications Memory Mapping page buffering, implicit 2P locking Callback 2 Phase Locking lock cache in the client C++ interface (ODMG like) ERSADS 95 Shared Distributed Memory : the Workspace Model 14

8 Futures Works Merge of WEA-YOODA Applications Document Database on heterogeneous networks Cooperative Information Systems possible extension to Asynchronous Video Stream ERSADS 95 Shared Distributed Memory : the Workspace Model 15 Mixing Data and Query Shipping Data Shipping moves confidential data to unsecured clients Query Shipping involves bigger servers back to the Mainframe Model Merging Data-Query Shipping = 2 kinds of access permissions 1 for operation on server 1 for data export consistency data between Client and Server ERSADS 95 Shared Distributed Memory : the Workspace Model