[
https://issues.apache.org/jira/browse/HBASE-2939?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12904950#action_12904950
]
Karthick Sankarachary commented on HBASE-2939:
----------------------------------------------
Ryan,
Thanks for the feedback - it is much appreciated. Some of the concerns you
raised should be addressed by the second version of the patch which:
* Modifies the "threadlocal" pool's #values method so that it return all
connections across all threads, instead of the one local to the current thread
that could potentially be null, which should resolve the NPEs on client
shutdown.
* Defaults the pool type to a bounded "roundrobin" of size 1, which would
emulate pre-patch behavior. On the other hand, the "threadlocal" pool's size is
bounded only by the clients' thread pool size.
Additionally, it:
* Renames the SharedMap class to PoolMap, for lack of a better name.
* Adds a test case for the PoolMap, which demonstrates how pools of different
types and sizes behave under different loads.
* Constrains the value for "hbase.client.ipc.pool.type" to one of the
following: "roundrobin", and "threadlocal".
To your point about file descriptor limits, the prescribed workaround for that
is described
[here|http://www.cloudera.com/blog/2009/03/configuration-parameters-what-can-you-just-ignore/].
We found out about that limit the hard way too.
The 20% speed up in "threadlocal" is good, although it may be even better if
the benchmark runs at high throughputs. The "roundrobin" pool should not be
unbounded, which is no longer the default - in general, it should be somewhere
between 1 and the number of user-defined threads.
> Allow Client-Side Connection Pooling
> ------------------------------------
>
> Key: HBASE-2939
> URL: https://issues.apache.org/jira/browse/HBASE-2939
> Project: HBase
> Issue Type: Improvement
> Components: client
> Affects Versions: 0.89.20100621
> Reporter: Karthick Sankarachary
> Attachments: HBASE-2939.patch, HBASE-2939.patch
>
>
> By design, the HBase RPC client multiplexes calls to a given region server
> (or the master for that matter) over a single socket, access to which is
> managed by a connection thread defined in the HBaseClient class. While this
> approach may suffice for most cases, it tends to break down in the context of
> a real-time, multi-threaded server, where latencies need to be lower and
> throughputs higher.
> In brief, the problem is that we dedicate one thread to handle all
> client-side reads and writes for a given server, which in turn forces them to
> share the same socket. As load increases, this is bound to serialize calls on
> the client-side. In particular, when the rate at which calls are submitted to
> the connection thread is greater than that at which the server responds, then
> some of those calls will inevitably end up sitting idle, just waiting their
> turn to go over the wire.
> In general, sharing sockets across multiple client threads is a good idea,
> but limiting the number of such sockets to one may be overly restrictive for
> certain cases. Here, we propose a way of defining multiple sockets per server
> endpoint, access to which may be managed through either a load-balancing or
> thread-local pool. To that end, we define the notion of a SharedMap, which
> maps a key to a resource pool, and supports both of those pool types.
> Specifically, we will apply that map in the HBaseClient, to associate
> multiple connection threads with each server endpoint (denoted by a
> connection id).
> Currently, the SharedMap supports the following types of pools:
> * A ThreadLocalPool, which represents a pool that builds on the
> ThreadLocal class. It essentially binds the resource to the thread from which
> it is accessed.
> * A ReusablePool, which represents a pool that builds on the LinkedList
> class. It essentially allows resources to be checked out, at which point it
> is (temporarily) removed from the pool. When the resource is no longer
> required, it should be returned to the pool in order to be reused.
> * A RoundRobinPool, which represents a pool that stores its resources in
> an ArrayList. It load-balances access to its resources by returning a
> different resource every time a given key is looked up.
> To control the type and size of the connection pools, we give the user a
> couple of parameters (viz. "hbase.client.ipc.pool.type" and
> "hbase.client.ipc.pool.size"). In case the size of the pool is set to a
> non-zero positive number, that is used to cap the number of resources that a
> pool may contain for any given key. A size of Integer#MAX_VALUE is
> interpreted to mean an unbounded pool.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
