Mailing-List: contact hdfs-issues-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Date: Wed, 24 Aug 2016 01:06:20 +0000 (UTC)
From: "Xiaoyu Yao (JIRA)" <jira@apache.org>
To: hdfs-issues@hadoop.apache.org
Message-ID: <JIRA.12992517.1469522930000.391180.1472000780579@Atlassian.JIRA>
In-Reply-To: <JIRA.12992517.1469522930000@Atlassian.JIRA>
References: <JIRA.12992517.1469522930000@Atlassian.JIRA> <JIRA.12992517.1469522930868@arcas>
Subject: [jira] [Commented] (HDFS-10690) Optimize insertion/removal of
 replica in ShortCircuitCache.java
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Wed, 24 Aug 2016 01:06:22 -0000


    [ https://issues.apache.org/jira/browse/HDFS-10690?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15433988#comment-15433988 ] 

Xiaoyu Yao commented on HDFS-10690:
-----------------------------------

bq.  I'll write a micro benchmark to compare LinkedMap with TreeMap and get back to you.

We have a lot of variations here: key, hash function, percentage of insert/delete/lookup operations, etc. There are quite some micro benchmark for TreeMap/LinkedHashMap(similar to LinkedMap from apache common) if you search online. Instead of tuning the parameter to simulate the workloads with micro benchmark, I would suggest we run the same YCSB benchmark with the last patch using LinkedMap for easy comparison. 

> Optimize insertion/removal of replica in ShortCircuitCache.java
> ---------------------------------------------------------------
>
>                 Key: HDFS-10690
>                 URL: https://issues.apache.org/jira/browse/HDFS-10690
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: hdfs-client
>    Affects Versions: 3.0.0-alpha2
>            Reporter: Fenghua Hu
>            Assignee: Fenghua Hu
>         Attachments: HDFS-10690.001.patch, HDFS-10690.002.patch, ShortCircuitCache_LinkedMap.patch
>
>   Original Estimate: 336h
>  Remaining Estimate: 336h
>
> Currently in ShortCircuitCache, two TreeMap objects are used to track the cached replicas.
> private final TreeMap<Long, ShortCircuitReplica> evictable = new TreeMap<>();
> private final TreeMap<Long, ShortCircuitReplica> evictableMmapped = new TreeMap<>();
> TreeMap employs Red-Black tree for sorting. This isn't an issue when using traditional HDD. But when using high-performance SSD/PCIe Flash, the cost inserting/removing an entry  becomes considerable.
> To mitigate it, we designed a new list-based for replica tracking.
> The list is a double-linked FIFO. FIFO is time-based, thus insertion is a very low cost operation. On the other hand, list is not lookup-friendly. To address this issue, we introduce two references into ShortCircuitReplica object.
> ShortCircuitReplica next = null;
> ShortCircuitReplica prev = null;
> In this way, lookup is not needed when removing a replica from the list. We only need to modify its predecessor's and successor's references in the lists.
> Our tests showed up to 15-50% performance improvement when using PCIe flash as storage media.
> The original patch is against 2.6.4, now I am porting to Hadoop trunk, and patch will be posted soon.


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org