openstack-manuals/doc/common/section_objectstorage-ringbuilder.xml
Christian Berendt 97334b6859 Remove unwanted unicode charaters
This patch is necessary because of I53e999fc91336871e1c32c70745f7d7cf2e256cf.

The following unicode characters will be removed:

* “...”
* ‘...’
* ― and —

Change-Id: If11a2d4ebd98b53f9f0d077b319983735f2e4b6b
2015-01-07 11:36:19 +01:00

227 lines
13 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<section xmlns="http://docbook.org/ns/docbook"
xmlns:xi="http://www.w3.org/2001/XInclude"
xmlns:xlink="http://www.w3.org/1999/xlink"
version="5.0"
xml:id="section_objectstorage-ringbuilder">
<title>Ring-builder</title>
<para>Use the swift-ring-builder utility to build and manage rings. This
utility assigns partitions to devices and writes an optimized
Python structure to a gzipped, serialized file on disk for
transmission to the servers. The server processes occasionally
check the modification time of the file and reload in-memory
copies of the ring structure as needed. If you use a slightly
older version of the ring, one of the three replicas for a
partition subset will be incorrect because of the way the
ring-builder manages changes to the ring. You can work around
this issue.</para>
<para>The ring-builder also keeps its own builder file with the
ring information and additional data required to build future
rings. It is very important to keep multiple backup copies of
these builder files. One option is to copy the builder files
out to every server while copying the ring files themselves.
Another is to upload the builder files into the cluster
itself. If you lose the builder file, you have to create a new
ring from scratch. Nearly all partitions would be assigned to
different devices and, therefore, nearly all of the stored
data would have to be replicated to new locations. So,
recovery from a builder file loss is possible, but data would
be unreachable for an extended time.</para>
<section xml:id="section_ring-data-structure">
<title>Ring data structure</title>
<para>The ring data structure consists of three top level
fields: a list of devices in the cluster, a list of lists
of device ids indicating partition to device assignments,
and an integer indicating the number of bits to shift an
MD5 hash to calculate the partition for the hash.</para>
</section>
<section xml:id="section_partition-assignment">
<title>Partition assignment list</title>
<para>This is a list of <literal>array('H')</literal> of
devices ids. The outermost list contains an
<literal>array('H')</literal> for each replica. Each
<literal>array('H')</literal> has a length equal to
the partition count for the ring. Each integer in the
<literal>array('H')</literal> is an index into the
above list of devices. The partition list is known
internally to the Ring class as
<literal>_replica2part2dev_id</literal>.</para>
<para>So, to create a list of device dictionaries assigned to
a partition, the Python code would look like:
<programlisting>devices = [self.devs[part2dev_id[partition]] for
part2dev_id in self._replica2part2dev_id]</programlisting></para>
<para>That code is a little simplistic because it does not
account for the removal of duplicate devices. If a ring
has more replicas than devices, a partition will have more
than one replica on a device.</para>
<para><literal>array('H')</literal> is used for memory
conservation as there may be millions of
partitions.</para>
</section>
<section xml:id="section_fractional-replicas">
<title>Replica counts</title>
<para>To support the gradual change in replica counts, a ring
can have a real number of replicas and is not restricted
to an integer number of replicas.</para>
<para>A fractional replica count is for the whole ring and not
for individual partitions. It indicates the average number
of replicas for each partition. For example, a replica
count of 3.2 means that 20 percent of partitions have four
replicas and 80 percent have three replicas.</para>
<para>The replica count is adjustable.</para>
<para>Example:</para>
<screen><prompt>$</prompt> <userinput>swift-ring-builder account.builder set_replicas 4</userinput>
<prompt>$</prompt> <userinput>swift-ring-builder account.builder rebalance</userinput></screen>
<para>You must rebalance the replica ring in globally
distributed clusters. Operators of these clusters
generally want an equal number of replicas and regions.
Therefore, when an operator adds or removes a region, the
operator adds or removes a replica. Removing unneeded
replicas saves on the cost of disks.</para>
<para>You can gradually increase the replica count at a rate
that does not adversely affect cluster performance.</para>
<para>For example:</para>
<screen><prompt>$</prompt> <userinput>swift-ring-builder object.builder set_replicas 3.01</userinput>
<prompt>$</prompt> <userinput>swift-ring-builder object.builder rebalance</userinput>
<computeroutput>&lt;distribute rings and wait&gt;...</computeroutput>
<prompt>$</prompt> <userinput>swift-ring-builder object.builder set_replicas 3.02</userinput>
<prompt>$</prompt> <userinput>swift-ring-builder object.builder rebalance</userinput>
<computeroutput>&lt;creatdistribute rings and wait&gt;...</computeroutput></screen>
<para>Changes take effect after the ring is rebalanced.
Therefore, if you intend to change from 3 replicas to 3.01
but you accidentally type <literal>2.01</literal>, no data
is lost.</para>
<para>Additionally, <command>swift-ring-builder
<replaceable>X.builder</replaceable>
create</command> can now take a decimal argument for
the number of replicas.</para>
</section>
<section xml:id="section_partition-shift-value">
<title>Partition shift value</title>
<para>The partition shift value is known internally to the
Ring class as <literal>_part_shift</literal>. This value
is used to shift an MD5 hash to calculate the partition
where the data for that hash should reside. Only the top
four bytes of the hash is used in this process. For
example, to compute the partition for the
<literal>/account/container/object</literal> path, the
Python code might look like the following code:
<programlisting>partition = unpack_from('&gt;I',
md5('/account/container/object').digest())[0] &gt;&gt;
self._part_shift</programlisting></para>
<para>For a ring generated with part_power P, the partition
shift value is <literal>32 - P</literal>.</para>
</section>
<section xml:id="section_build-ring">
<title>Build the ring</title>
<para>The ring builder process includes these high-level
steps:</para>
<orderedlist>
<listitem>
<para>The utility calculates the number of partitions to
assign to each device based on the weight of the
device. For example, for a partition at the power
of 20, the ring has 1,048,576 partitions. One
thousand devices of equal weight each want
1,048.576 partitions. The devices are sorted by
the number of partitions they desire and kept in
order throughout the initialization
process.</para>
<note>
<para>Each device is also assigned a random
tiebreaker value that is used when two devices
desire the same number of partitions. This
tiebreaker is not stored on disk anywhere, and
so two different rings created with the same
parameters will have different partition
assignments. For repeatable partition
assignments,
<literal>RingBuilder.rebalance()</literal>
takes an optional seed value that seeds the
Python pseudo-random number generator.</para>
</note>
</listitem>
<listitem>
<para>The ring builder assigns each partition replica
to the device that requires most partitions at
that point while keeping it as far away as
possible from other replicas. The ring builder
prefers to assign a replica to a device in a
region that does not already have a replica. If no
such region is available, the ring builder
searches for a device in a different zone, or on a
different server. If it does not find one, it
looks for a device with no replicas. Finally, if
all options are exhausted, the ring builder
assigns the replica to the device that has the
fewest replicas already assigned.</para>
<note>
<para>The ring builder assigns multiple replicas
to one device only if the ring has fewer
devices than it has replicas.</para>
</note>
</listitem>
<listitem>
<para>When building a new ring from an old ring, the
ring builder recalculates the desired number of
partitions that each device wants.</para>
</listitem>
<listitem>
<para>The ring builder unassigns partitions and
gathers these partitions for reassignment, as
follows: <itemizedlist>
<listitem>
<para>The ring builder unassigns any
assigned partitions from any removed
devices and adds these partitions to
the gathered list.</para>
</listitem>
<listitem>
<para>The ring builder unassigns any
partition replicas that can be spread
out for better durability and adds
these partitions to the gathered list.
</para>
</listitem>
<listitem>
<para>The ring builder unassigns random
partitions from any devices that have
more partitions than they need and
adds these partitions to the gathered
list.</para>
</listitem>
</itemizedlist></para>
</listitem>
<listitem>
<para>The ring builder reassigns the gathered
partitions to devices by using a similar method to
the one described previously.</para>
</listitem>
<listitem>
<para>When the ring builder reassigns a replica to a
partition, the ring builder records the time of
the reassignment. The ring builder uses this value
when it gathers partitions for reassignment so
that no partition is moved twice in a configurable
amount of time. The RingBuilder class knows this
configurable amount of time as
<literal>min_part_hours</literal>. The ring
builder ignores this restriction for replicas of
partitions on removed devices because removal of a
device happens on device failure only, and
reassignment is the only choice.</para>
</listitem>
</orderedlist>
<para>Theses steps do not always perfectly rebalance a ring
due to the random nature of gathering partitions for
reassignment. To help reach a more balanced ring, the
rebalance process is repeated until near perfect (less
than 1 percent off) or when the balance does not improve
by at least 1 percent (indicating we probably cannot get
perfect balance due to wildly imbalanced zones or too many
partitions recently moved).</para>
</section>
</section>