Archive for the ‘Zing JVM’ Category

Lucene with Zing, Part 2

Wednesday, November 21st, 2012

Lucene with Zing, Part 2 by Mike McCandless.

From the post:

When I last tested Lucene with the Zing JVM the results were impressive: Zing’s fully concurrent C4 garbage collector had very low pause times with the full English Wikipedia index (78 GB) loaded into RAMDirectory, which is not an easy feat since we know RAMDirectory is stressful for the garbage collector.

I had used Lucene 4.0.0 alpha for that first test, so I decided to re-test using Lucene’s 4.0.0 GA release and, surprisingly, the results changed! MMapDirectory’s max throughput was now better than RAMDirectory’s (versus being much lower before), and the concurrent mark/sweep collector (-XX:-UseConcMarkSweepGC) was no longer hitting long GC pauses.

This was very interesting! What change could improve MMapDirectory’s performance, and lower the pressure on concurrent mark/sweep’s GC to the point where pause times were so much lower in GA compared to alpha?

Mike updates his prior experience with Lucene and Zing.

Covers the use gcLogAnalyser and Fragger to understand “why” his performance test results changed from the alpha to GA releases.

Insights into both Lucene and Zing.

Have you considered loading your topic map into RAM?

Indexes in RAM?

Wednesday, August 1st, 2012

The Mike McCandless post: Lucene index in RAM with Azul’s Zing JVM will help make your case for putting your index in RAM!

From the post:

Google’s entire index has been in RAM for at least 5 years now. Why not do the same with an Apache Lucene search index?

RAM has become very affordable recently, so for high-traffic sites the performance gains from holding the entire index in RAM should quickly pay for the up-front hardware cost.

The obvious approach is to load the index into Lucene’s RAMDirectory, right?

Unfortunately, this class is known to put a heavy load on the garbage collector (GC): each file is naively held as a List of byte[1024] fragments (there are open Jira issues to address this but they haven’t been committed yet). It also has unnecessary synchronization. If the application is updating the index (not just searching), another challenge is how to persist ongoing changes from RAMDirectory back to disk. Startup is much slower as the index must first be loaded into RAM. Given these problems, Lucene developers generally recommend using RAMDirectory only for small indices or for testing purposes, and otherwise trusting the operating system to manage RAM by using MMapDirectory (see Uwe’s excellent post for more details).

While there are open issues to improve RAMDirectory (LUCENE-4123 and LUCENE-3659), they haven’t been committed and many users simply use RAMDirectory anyway.

Recently I heard about the Zing JVM, from Azul, which provides a pauseless garbage collector even for very large heaps. In theory the high GC load of RAMDirectory should not be a problem for Zing. Let’s test it! But first, a quick digression on the importance of measuring search response time of all requests.

There are obvious speed advantages to holding indexes in RAM.

Curious, is RAM just a quick disk? Or do we need to think about data structures/access differently with RAM? Pointers?