IBM Parallel Sysplex
In computing, a Parallel Sysplex is a cluster of IBM mainframes acting together as
a single system image with z/OS. Used for disaster recovery, Parallel Sysplex combines
data sharing and parallel computing to allow a cluster of up to 32 systems to share a
workload for high performance and high availability.
Sysplex[edit]
In 1990, IBM mainframe computers introduced the concept of a Systems Complex,
commonly called a Sysplex, with MVS/ESA SPV4.1. This allows authorized
components in up to eight logical partitions (LPARs) to communicate and cooperate
with each other using the XCF protocol.
Components of a Sysplex include:
A common time source to synchronize all member systems' clocks. This can
involve either a Sysplex timer (Model 9037), or the Server Time Protocol (STP)
Global Resource Serialization (GRS), which allows multiple systems to access
the same resources concurrently, serializing where necessary to ensure exclusive
access
Cross System Coupling Facility (XCF), which allows systems to
communicate peer-to-peer
Couple Data Sets (CDS)
Users of a (base) Sysplex include:
Console services – allowing one to merge multiple MCS consoles from the
different members of the Sysplex, providing a single system image for Operations
Automatic Restart Manager (ARM) – Policy to direct automatic restart of failed
jobs or started tasks on the same system if it is available or on another LPAR in the
Sysplex
Sysplex Failure Manager (SFM) – Policy that specifies automated actions to take
when certain failures occur such as loss of a member of a Sysplex or when
reconfiguring systems
Workload Manager (WLM) – Policy based performance management of
heterogeneous workloads across one or more z/OS images or even on AIX
Global Resource Serialization (GRS) - Communication – allows use of XCF links
instead of dedicated channels for GRS, and Dynamic RNLs
Tivoli OPC – Hot standby support for the controller
RACF (IBM's mainframe security software product) – Sysplex-wide RVARY and
SETROPTS commands
PDSE file sharing
Multisystem VLFNOTE, SDUMP, SLIP, DAE
Resource Measurement Facility (RMF) – Sysplex-wide reporting
CICS – uses XCF to provide better performance and response time than using
VTAM for transaction routing and function shipping.
zFS – Using XCF communication to access data across multiple LPARs
Parallel Sysplex[edit]
Schematic representation of a Parallel Sysplex
The Parallel Sysplex was introduced with the addition of the Coupling Facility (CF) with
coupling links for high speed communication, with MVS/ESA V5.1 operating system
support, together with the mainframe models in April 1994. [1]
The Coupling Facility (CF) may reside on a dedicated stand-alone server configured
with processors that can run Coupling Facility control code (CFCC), as integral
processors on the mainframes themselves configured as ICFs (Internal Coupling
Facilities), or less common, as normal LPARs. The CF contains Lock, List, and Cache
structures to help with serialization, message passing, and buffer consistency between
multiple LPARs.[2]
The primary goal of a Parallel Sysplex is to provide data sharing capabilities, allowing
multiple databases for direct reads and writes to shared data. This can provide benefits
of
Help remove single points of failure within the server, LPAR, or subsystems
Application Availability
Single System Image
Dynamic Session Balancing
Dynamic Transaction Routing
Scalable capacity
Databases running on the System z server that can take advantage of this include:
DB2
IBM Information Management System (IMS).
VSAM (VSARM/RLS)
IDMS
Adabas
DataCom
Oracle
Other components can use the Coupling Facility to help with system management,
performance, or reduced hardware requirements. Called “Resource Sharing”, uses
include:
Catalog – shared catalogs to improve performance by reducing I/O to a catalog
data set on disk
CICS – Using the CF to provide sharing and recovery capabilities for named
counters, data tables, or transient data
DFSMShsm – Workload balancing for data migration workload
GRS Star – Reduced CPU and response time performance for data set
allocation.
Tape Switching uses the GRS structure to provide sharing of tape units between z/OS
images.
Dynamic CHPID Management (DCM), and I/O priority management
JES2 Checkpoint – Provides improved access to a multisystem checkpoint
Operlog / Logrec – Merged multisystem logs for system management
RACF – shared data set to simplify security management across the Parallel
Sysplex
WebSphere MQ – Shared message queues for availability and flexibility
WLM - provides support for Intelligent Resource Director (IRD) to extends the
z/OS Workload Manager to help manage CPU and I/O resources across multiple
LPARs within the Parallel Sysplex. Functions include LPAR CPU management, IRD.
Multi-system enclave management for improved performance
XCF Star – Reduced hardware requirements and simplified management of XCF
communication paths
Major components of a Parallel Sysplex include:
Coupling Facility (CF or ICF) hardware, allowing multiple processors to share,
cache, update, and balance data access;
Sysplex Timers or Server Time Protocol to synchronize the clocks of all member
systems;
High speed, high quality, redundant cabling;
Software (operating system services and, usually, middleware such as DB2).
The Coupling Facility may be either a dedicated external system (a small mainframe,
such as a System z9 BC, specially configured with only coupling facility processors) or
integral processors on the mainframes themselves configured as ICFs (Internal
Coupling Facilities).[3] It is recommended that at least one external CF be used in a
parallel sysplex.[4] It is recommended that a Parallel Sysplex has at least two CFs and/or
ICFs for redundancy, especially in a production data sharing environment. Server Time
Protocol (STP) replaced the Sysplex Timers beginning in 2005 for System z mainframe
models z990 and newer.[5] A Sysplex Timer is a physically separate piece of hardware
from the mainframe,[6] whereas STP is an integral facility within the mainframe's
microcode.[7] With STP and ICFs it is possible to construct a complete Parallel Sysplex
installation with two connected mainframes. Moreover, a single mainframe can contain
the internal equivalent of a complete physical Parallel Sysplex, useful for application
testing and development purposes.[8]
The IBM Systems Journal dedicated a full issue to all the technology components. [9]
