ibm-information-center/dist/eclipse/plugins/i5OS.ic.rzaly_5.4.0.1/rzalyswitchablevsstandalone.htm

129 lines
8.6 KiB
HTML
Raw Normal View History

2024-04-02 14:02:31 +00:00
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html lang="en-us" xml:lang="en-us">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="security" content="public" />
<meta name="Robots" content="index,follow" />
<meta http-equiv="PICS-Label" content='(PICS-1.1 "http://www.icra.org/ratingsv02.html" l gen true r (cz 1 lz 1 nz 1 oz 1 vz 1) "http://www.rsac.org/ratingsv01.html" l gen true r (n 0 s 0 v 0 l 0) "http://www.classify.org/safesurf/" l gen true r (SS~~000 1))' />
<meta name="DC.Type" content="topic" />
<meta name="DC.Title" content="Switchable and stand-alone independent disk pools" />
<meta name="DC.Relation" scheme="URI" content="rzalyconcepts.htm" />
<meta name="copyright" content="(C) Copyright IBM Corporation 2002, 2006" />
<meta name="DC.Rights.Owner" content="(C) Copyright IBM Corporation 2002, 2006" />
<meta name="DC.Format" content="XHTML" />
<meta name="DC.Identifier" content="rzalyswitchablevsstandalone" />
<meta name="DC.Language" content="en-us" />
<!-- All rights reserved. Licensed Materials Property of IBM -->
<!-- US Government Users Restricted Rights -->
<!-- Use, duplication or disclosure restricted by -->
<!-- GSA ADP Schedule Contract with IBM Corp. -->
<link rel="stylesheet" type="text/css" href="./ibmdita.css" />
<link rel="stylesheet" type="text/css" href="./ic.css" />
<title>Switchable and stand-alone independent disk pools</title>
</head>
<body id="rzalyswitchablevsstandalone"><a name="rzalyswitchablevsstandalone"><!-- --></a>
<!-- Java sync-link --><script language="Javascript" src="../rzahg/synch.js" type="text/javascript"></script>
<h1 class="topictitle1">Switchable and stand-alone independent disk pools</h1>
<div><p>There are two basic environments in which you can take advantage of independent
disk pools: a multisystem environment managed by an iSeries™ cluster, and a single-system
environment with a single iSeries server.</p>
<div class="section"><h4 class="sectiontitle">Independent disk pools in a multisystem clustered environment</h4><div class="p"><img src="./delta.gif" alt="Start of change" />A group of servers in a cluster can take advantage of the switchover
capability to move access to the independent disk pool from server to server.
In this environment, an independent disk pool can be switchable when it resides
on a switchable device. A switchable device can be an external expansion unit
(tower), an input/output processor (IOP) on the bus shared by logical partitions,
or an IOP or IOPless hardware that is assigned to an I/O pool. <div class="note"><span class="notetitle">Note:</span> Hardware
that does not have a physical IOP has a virtual logical representation of
the IOP.</div>
A switchable device that contains an independent disk pool
can be switched automatically in the case of an unplanned outage, or it can
be switched manually by administering a switchover.<img src="./deltaend.gif" alt="End of change" /></div>
<p>Another
option that can be leveraged in a multisystem environment is <a href="rzalygeographicmirror.htm">geographic mirroring</a>. Geographic mirroring allows you
to maintain two identical copies of an independent disk pool at two sites
that are geographically separated. The independent disk pools at the separate
sites can be switchable or dedicated.</p>
</div>
<div class="section"><h4 class="sectiontitle">Dedicated independent disk pools in a single-system environment</h4>An
independent disk pool in a single-system environment, with no clustering
and no switchable devices, is said to be a dedicated, private, or stand-alone
independent disk pool. While you cannot switch the access to the independent
disk pool amongst servers in this environment, you can still isolate data
in an independent disk pool, keeping it separate from the rest of the disk
storage on the server. The independent disk pool can then be made available
(brought online) and made unavailable (taken offline) as needed. This might
be done, for example, to isolate data associated with a specific application
program or to isolate low-use data that is only needed periodically. Dedicated
independent disk pools might also be used to consolidate data from several
small servers at branch offices to one or more larger servers at a central
location, while still keeping the data separate for each branch. <p>Independent
disk pools allow you to isolate certain maintenance functions. Then, when
you need to perform disk management functions that normally require the entire
system to be at DST, you can perform them by merely varying off the affected
independent disk pool.</p>
<p>The following table compares dedicated independent
disk pools and independent disk pools in a multisystem environment.</p>
<div class="tablenoborder"><table cellpadding="4" cellspacing="0" summary="" width="100%" frame="border" border="1" rules="all"><thead align="left"><tr class="tablemainheaderbar" valign="bottom"><th class="oddrowblue" rowspan="2" valign="bottom" width="25.909528392685278%" id="d0e47">Consideration  </th>
<th class="oddrowblue" valign="bottom" width="26.564003849855634%" id="d0e49"> Dedicated</th>
<th class="oddrowblue" colspan="2" align="center" valign="middle" id="d0e51"> Multisystem environment</th>
</tr>
<tr valign="bottom"><th class="oddrowblue" valign="bottom" width="26.564003849855634%" id="d0e54"> Single system</th>
<th class="oddrowblue" valign="bottom" width="25.293551491819056%" id="d0e56">Multisystem cluster</th>
<th class="oddrowblue" valign="bottom" width="22.23291626564004%" id="d0e58">Logical partitions in a cluster</th>
</tr>
</thead>
<tbody><tr><td valign="top" width="25.909528392685278%" headers="d0e47 ">iSeries cluster
required</td>
<td valign="top" width="26.564003849855634%" headers="d0e49 d0e54 ">No</td>
<td valign="top" width="25.293551491819056%" headers="d0e51 d0e56 ">Yes</td>
<td valign="top" width="22.23291626564004%" headers="d0e51 d0e58 "> Yes</td>
</tr>
<tr><td valign="top" width="25.909528392685278%" headers="d0e47 ">Connectivity between systems</td>
<td valign="top" width="26.564003849855634%" headers="d0e49 d0e54 ">Not applicable</td>
<td valign="top" width="25.293551491819056%" headers="d0e51 d0e56 ">HSL loop</td>
<td valign="top" width="22.23291626564004%" headers="d0e51 d0e58 ">Virtual OptiConnect</td>
</tr>
<tr><td valign="top" width="25.909528392685278%" headers="d0e47 ">Location of disk units</td>
<td valign="top" width="26.564003849855634%" headers="d0e49 d0e54 ">Any supported internal or external disk units</td>
<td valign="top" width="25.293551491819056%" headers="d0e51 d0e56 ">External expansion unit (tower)</td>
<td valign="top" width="22.23291626564004%" headers="d0e51 d0e58 ">IOP on shared bus</td>
</tr>
<tr><td valign="top" width="25.909528392685278%" headers="d0e47 ">Switchability</td>
<td valign="top" width="26.564003849855634%" headers="d0e49 d0e54 ">No</td>
<td valign="top" width="25.293551491819056%" headers="d0e51 d0e56 ">Yes, between systems</td>
<td valign="top" width="22.23291626564004%" headers="d0e51 d0e58 ">Yes, between partitions</td>
</tr>
<tr><td valign="top" width="25.909528392685278%" headers="d0e47 ">Switchable entity</td>
<td valign="top" width="26.564003849855634%" headers="d0e49 d0e54 ">None</td>
<td valign="top" width="25.293551491819056%" headers="d0e51 d0e56 "> Expansion unit</td>
<td valign="top" width="22.23291626564004%" headers="d0e51 d0e58 "> IOP</td>
</tr>
</tbody>
</table>
</div>
<p>In a hardware switching environment, one node in the device domain
owns it, and all the other nodes in the device domain show that the independent
disk pool exists. In a geographic mirroring environment, one node at each
site owns a copy of the independent disk pool. When an independent disk pool
is created or deleted, the node creating or deleting the independent disk
pool informs all the other nodes in the device domain of the change. If clustering
is not active between the nodes, or a node is in the midst of a long running
disk pool configuration change, that node will not update and will be inconsistent
with the node rest of the nodes. Nodes must be consistent before a failover
or switchover. Ending clustering and starting clustering will ensure that
the configuration is consistent. </p>
<p>For more on switchable and dedicated
independent disk pools, including example configurations for each of these
environments, see <a href="rzalyscenarios.htm">Examples: Independent disk pool configurations</a>.</p>
</div>
</div>
<div>
<div class="familylinks">
<div class="parentlink"><strong>Parent topic:</strong> <a href="rzalyconcepts.htm">Independent disk pools</a></div>
</div>
</div>
</body>
</html>