Server system architecture

SERVER SYSTEM ARCHITECTURE
 Server systems can be broadly categorized into two kinds:
 transaction servers which are widely used in relational database
systems, and
 data servers, used in object-oriented database systems

TRANSACTION SERVERS
 Also called query server systems or SQL server systems
o Clients send requests to the server
o Transactions are executed at the server
o Results are shipped back to the client.
 Requests are specified in SQL, and communicated to the server through a remote
procedure call (RPC) mechanism.
 Transactional RPC allows many RPC calls to form a transaction.
 Open Database Connectivity (ODBC) is a C language application program interface
standard from Microsoft for connecting to a server, sending SQL requests, and receiving
results.
 JDBC standard is similar to ODBC, for Java

TRANSACTION SERVER PROCESS STRUCTURE
 A typical transaction server consists of multiple processes accessing data in shared
memory.
 Server processes
o These receive user queries (transactions), execute them and send results back
o Processes may be multithreaded, allowing a single process to execute several user
queries concurrently
o Typically multiple multithreaded server processes
 Lock manager process
o More on this later
 Database writer process
o Output modified buffer blocks to disks continually

TRANSACTION SERVER PROCESSES
(CONT.)
 Log writer process
o Server processes simply add log records to log record buffer
o Log writer process outputs log records to stable storage.
 Checkpoint process
o Performs periodic checkpoints
 Process monitor process
o Monitors other processes, and takes recovery actions if any of the other processes fail
 E.g., aborting any transactions being executed by a server process and restarting it

TRANSACTION SYSTEM PROCESSES (CONT.)

TRANSACTION SYSTEM PROCESSES (CONT.)
 Shared memory contains shared data
o Buffer pool
o Lock table
o Log buffer
o Cached query plans (reused if same query submitted again)
 All database processes can access shared memory
 To ensure that no two processes are accessing the same data structure at the same time,
databases systems implement mutual exclusion using either
o Operating system semaphores
o Atomic instructions such as test-and-set
 To avoid overhead of inter process communication for lock request/grant, each database
process operates directly on the lock table
o instead of sending requests to lock manager process
 Lock manager process still used for deadlock detection

DATA SERVERS
 Used in high-speed LANs, in cases where
 The clients are comparable in processing power to the server
 The tasks to be executed are compute intensive.
 Data are shipped to clients where processing is performed, and then shipped results
back to the server.
 This architecture requires full back-end functionality at the clients.
 Used in many object-oriented database systems
 Issues:
 Page-Shipping versus Item-Shipping
 Locking
 Data Caching
 Lock Caching

DATA SERVERS (CONT.)
 Page-shipping versus item-shipping
 Smaller unit of shipping  more messages
 Worth prefetching related items along with requested item
 Page shipping can be thought of as a form of prefetching
 Locking
 Overhead of requesting and getting locks from server is high due to message delays
 Can grant locks on requested and prefetched items; with page shipping, transaction is
granted lock on whole page.
 Locks on a prefetched item can be P{called back} by the server, and returned by client
transaction if the prefetched item has not been used.
 Locks on the page can be deescalated to locks on items in the page when there are lock
conflicts. Locks on unused items can then be returned to server.

DATA SERVERS (CONT.)
 Data Caching
 Data can be cached at client even in between transactions
 But check that data is up-to-date before it is used (cache coherency)
 Check can be done when requesting lock on data item
 Lock Caching
 Locks can be retained by client system even in between transactions
 Transactions can acquire cached locks locally, without contacting server
 Server calls back locks from clients when it receives conflicting lock request. Client
returns lock once no local transaction is using it.
 Similar to de-escalation, but across transactions.

Server system architecture

Recommended

More Related Content

What's hot (20)

Similar to Server system architecture (20)

Recently uploaded (20)

Server system architecture