<html><head></head><body>elasticsearch version <code>0.20.2</code> is out. This release includes both important bug fixes as well as upgrades to Lucene (from 3.6.1 to 3.6.2), and other internal libraries.
Another change includes how plugins are downloaded. Now, &#x201C;elasticsearch&#x201D; plugins are downloaded automatically from our new <code>download.elasticsearch.org</code> service, but it can also download plugins from maven directly. Site plugins can stil be downloaded from github as &#x201C;repositories&#x201D;. In the future, we will also allow users to upload their own plugins to <code>download.elasticsearch.org</code> if they wish to.
Enjoy!
</body></html>

0.20.2 Released

<html><head></head><body>Elasticsearch has been using Github&#x2019;s static downloads feature as a means of distributing releases for both the core and its plugins for a few years now. Unfortunately, last week, <a href="https://github.com/blog/1302-goodbye-uploads">Github stopped supporting downloads</a>. We&#x2019;ve been working the past few days to replace this functionality internally.
All of the links to elasticsearch core have been updated to the new site through <code>download.elasticsearch.org</code>. We&#x2019;re still migrating our plugins, so you&#x2019;ll see those change in the next day or two. We are still working on a solution for third-party plugins.
The old assets will be removed tomorrow, so if you&#x2019;re using any <code>github.com/downloads/elasticsearch/*</code> links in your provisioning system please update them to the commensurate <code>download.elasticsearch.org/elasticsearch/*</code>.
</body></html>

New Download Service

<html><head></head><body>elasticsearch version <code>0.20.0</code> is out. This is the GA release of the 0.20.x branch. You can download it <a href="/download">here</a>. The release mainly include bug fixes over the previous <code>0.20.0.RC1</code> release.
Update: <code>0.20.1</code> released to fix a bug when downloading plugins, please make sure to use it when using plugins.
This version also deprecates the snappy compression option (see more <a href="https://github.com/elasticsearch/elasticsearch/issues/2459">here</a>), and the Shared Gateway (more info <a href="https://github.com/elasticsearch/elasticsearch/issues/2458">here</a>).
</body></html>

0.20.0 Released

<html><head></head><body>elasticsearch version <code>0.19.12</code> is out. You can download it <a href="/download">here</a>. It contains several bug fixes (upgrade when possible&#x2026;) and a new feature for indexing slow log (similar to search slow log feature).
This version also deprecates the snappy compression option (see more <a href="https://github.com/elasticsearch/elasticsearch/issues/2459">here</a>), and the Shared Gateway (more info <a href="https://github.com/elasticsearch/elasticsearch/issues/2458">here</a>).
</body></html>

0.19.12 Released

Can't make it? Register and we'll send you the recording. You'll also receive an email with related content

Can't make it? Register and we'll send you the recording. You'll also receive an email with related content.

You'll also receive an email with related content.

You'll also receive an email with related content

Load More

By submitting you acknowledge that you've read and agree to our <a href="/legal/elastic-cloud-account-terms" target="_blank">Terms of Service</a>, and that Elastic may <a href="/legal/privacy-statement#how-we-use-the-information" target="_blank">contact you</a> about our related products and services, using the details you provide above. See <a href="/legal/privacy-statement/" target="_blank">Elastic’s Privacy Statement</a> for more details or to opt-out at any time.

Newsletter GDPR Text

Speakers

Close

See more insights

The content on this page is not available in the selected language. As Elastic grows globally, we continue to support content in multiple languages.

The content on this page is not available in the selected language.

Author

Learn more

Watch now (no PT)

Watch now

See all top stories

All (no PT translation)

Contact information

Press Release

Share on Reddit

Articles by

Share this story

Share by Email

Read less

Search Integrations

All Solutions

Thank you for registering. We will send you a confirmation email soon.

Thank you for your interest!

Follow us on Youtube

Follow us on Twitter

Follow us on LinkedIn

Follow us on Facebook

Headshot of

Reset all

Filters

Global Virtual Event

View more posts

Print

Continue reading

Share on Facebook

Share on LinkedIn

Share on Twitter

Share

Small image for

Video for

Explore similar demos

Register now

View next

Upcoming webinar

On-demand webinar

Featured webinar

Highlights

See when this webinar starts in my time zone

Related workshops

Hosted by

Agenda

Location

Date

More

View more learning opportunities

Load more press releases

Load more news

What to explore next...

See All Posts

Contact Info

Register to Watch

Sign In to Attend

Register to Attend

Welcome Clinton!

<html><head></head><body>With unbelievable excitement we just announced Elasticsearch secured $10M in Series A venture funding from <a href="http://www.benchmark.com" target="_blank">Benchmark Capital</a>, led by Peter Fenton. Other investors in the round include my SpringSource co-founder Rod Johnson and <a href="http://www.dcvc.com" target="_blank">Data Collective</a>. Being able to work with Peter Fenton and Rod Johnson again after SpringSource is nothing short of a privilege. Both are established open source visionaries and have already brought a lot of value to the company in terms of insights, network and general advice over the last few weeks.Three reasons stand out why Benchmark, Rod, and Data Collective believe in Elasticsearch: the product and the company:<ol>
<li>Adoption: 1.6 Million Downloads and Counting 
This is one of the areas in which Elasticsearch is special as an open source project. The project is currently being downloaded at a run rate of approximately 200,000 downloads per month, making it one of the most rapidly adopted solutions in big data and is used in mission critical applications by thousands of companies worldwide.</li>
<li>The Solution: Searching and Analyzing Big Data in Real Time 
Shay specifically designed Elasticsearch from the ground up to be a runtime environment that allows people to analyze and act on massive volumes of data &#x2013; in real time. Recent research from The Economist Intelligence Unit lists this exact objective to be the #1 challenge reported by their research population. Elasticsearch makes tackling this challenge easy and doesn&apos;t require you to be a data scientist to put it to good use.</li>
<li>Our Team: Deep Technical and Open Source Business Experience 
In the end our company is only going to be successful if the folks that are part of it are true professionals that know what they&apos;re doing, and have a laser-like focus on the success of anybody who uses Elasticsearch. This is what our team of former SpringSource veterans and technology specialist are doing every day.</li>
</ol>When Shay and I joined forces about 6 months ago, we set out to build a company based on our shared belief that we would be able to create a beacon of light in our industry. Our industry, mostly referred to as the big data market, is undeniably complex and extraordinary technology heavy. Many companies provide a solution that solves one of the many technical challenges companies are facing when aiming to gain competitive advantage from the massive volumes of data it&apos;s gathering on a daily basis. I believe that up until Shay created Elasticsearch, very few &#x2013; if any &#x2013;of the solutions available were able to make achieving this objective as easy as operating an Apple device.So in a way Elasticsearch already achieved its objective, which makes my life as the CEO of Elasticsearch a lot easier of course. However, we&apos;re not done. In fact, we&apos;re only getting started&#x2026;Since we launched the company, I have received many questions on where we want to take the company and what we plan on doing moving forward. The key element to the answer to that question has always been, that our way forward is merely a continuation of the path we&apos;re currently on. Please allow me to elaborate.The more we talk to people who use Elasticsearch, the more we are able to distill 3 main reasons why Elasticsearch is so popular. In a nutshell, people love our product because it is:<ol>
<li>Easy to use</li>
<li>Robust</li>
<li>Complete</li>
</ol>To us, this is an extremely important finding. Predominantly because this confirms that Shay and his team are achieving their objective: to make an extremely easy to use product that reliably delivers on its promise. The mere fact that a product achieve it&apos;s objective is actually not as common as it might seem. As products mature, markets mature and as product&apos;s established user base grows, it&apos;s not uncommon for product roadmaps to deviate from its original path due to heavy pressure from many outside forces. I believe that Elasticsearch was created after a solid vision, rather than based on feature requests from individual users. This has had a massive positive impact on Shay&apos;s ability to create a product that is usable for an immensely large audience, instead of just a handful of power users. Shay has not given in to temptations, but rather worked closely together with many users to create a user-friendly product that meets all critical requirements in a large number of use cases.The challenge for our company is to be as focused on being as easy to work with, robust and complete as our product is. Now that doesn&apos;t mean that there&apos;s nothing left to be desired. Au contraire, we are currently working hard on bullet proofing our support organization by scaling it out fast and fine-tuning our systems to ensure the quality of our support increases while our client base rapidly grows. We are also working on a public training course schedule that covers more cities across the globe, to ensure attendees never have to travel unacceptably far to attend a course. And as is to be expected, we are growing our executive team with talented professional that have the required skills to help grow our reach.With Elasticsearch now having a company behind it to propel it forward, it seems safe to conclude that the Big Data Dark Ages are now finally behind us.Steven &amp; Shay</body></html>

Elasticsearch: Easy, Complete and Robust – Just like Our Product

<html><head></head><body>I am very happy to announce that Igor Motov (<a href="http://twitter.com/imotov" target="_blank">&quot;imotov&quot;</a>) is joining Elasticsearch. Igor has extensive knowledge on Elasticsearch, and has written several plugins for it, including a zookeeper discovery module and the jetty plugin. He is also the co-organizer of the <a href="https://twitter.com/ElasticsearchMA" "="" target="_blank">&quot;Boston Elasticsearch meetup&quot;</a>. Igor will join our team focusing on continuing and improving Elasticsearch as his full time job.Personally, I am very excited about having Igor join us. We just got a big boost on our ability to execute on our vision that is Elasticsearch. Welcome Igor!</body></html>

Welcome Igor

<html><head></head><body>elasticsearch version <code>0.19.11</code> and <code>0.20.0.RC1</code> are out. You can download it <a href="/download">here</a>. <code>0.19.11</code> contains important bug fixes and is a recommended upgrade, <code>0.20.0.RC1</code> is the first release candidate for 0.20 release, major features include:
<h2>Update API</h2>
Improved update API, including the ability to provide a partial document to merge into the existing one, and the ability to &#x201C;upsert&#x201D; a document if the document doesn&#x2019;t exists.
<h2>geo_shape</h2>
Support for a <code>geo_shape</code> type allowing to index complex geo shape (using geo json as the format), and for a <code>geo_shape</code> query/filter allowing to provide a geo shape and check if it intersects, contains, or disjoint with the indexed shape. It also allows to fetch the &#x201C;query&#x201D; shape from the index itself (without needing to provide it as part of the query, just the name of the index/type/id to use).
<h2>Warmers</h2>
Provides the ability to register search requests that will be executed before new data becomes available for search, allowing to preload for example data that is needed for sorting, faceting, or parent/child support. This removes the load time from the &#x201C;user&#x201D; execution path, allowing to provide consistent search performance.
Warmers have full API control for registering them, as well as deleting and getting them. They can also be provided as part of the index creation API, or as part of index templates.
<h2>Other</h2>
A lot of work has going into building the basis for cross major version cluster compatibility, we are not there yet, but we are on our way. Bugs have been fixed (though most major ones have been backported to 0.19), and many other smaller features and enhancements have been implemented.
Moving forward, the plan is to have the next version be 0.20.0 final. 0.21 will focus on upgrading to Lucene 4.0 (and exposing its new features), as well as internal refactoring to support future features such as grouping (specifically, optimized for distributed execution).
</body></html>

0.19.11 & 0.20.0.RC1 Released

<html><head></head><body>elasticsearch version <code>0.19.10</code> is out. You can download it <a href="/download">here</a>. It contains several bug fixes but also new features, including:
<h2>has_parent filter/query</h2>
<code>has_parent</code> filter allows to execute a query on parent documents and result in matching children documents (the inverse of <code>has_child</code>). More info <a href="https://github.com/elasticsearch/elasticsearch/issues/2254">here</a>. Also, the performance of <code>has_child</code> filter/query has been greatly improved.
<h2>Less overhead of filter cache</h2>
The filter cache cleanup process is now more optimized, specially when large number of shards exists on a node.
<h2>Cluster Reroute API</h2>
The <a href="https://github.com/elasticsearch/elasticsearch/issues/2256">cluster reroute API</a> allows to explicitly send commands that would move, cancel, and allocate shards. Note, this is an experimental API and feedback is welcomed.
<h2>Bulk UDP</h2>
The <a href="https://github.com/elasticsearch/elasticsearch/issues/2201">Bulk UDP</a> interface allows to have elasticsearch listen using UDP for bulk formatted requests. While UDP comes with its own downsides, it can come in handy, specially when indexing non critical data (ala statsd). This interface is also experimental.
</body></html>

0.19.10 Released

<html><head></head><body>I very happy to announce that Karel Minarik (<a href="https://github.com/karmi">&quot;karmi&quot;</a>) is joining Elasticsearch. Karel is well known in the Elasticsearch community, specifically the ruby community as the author of the popular <a href="https://github.com/karmi/tire">tire</a> library, though he is quite the polyglot that dabbles in chef, emberjs and many more.Go ahead and read <a href="http://www.elasticsearch.com/blog/the-future-is-elastic/" title="The future is elastic">Karel thoughts on joining Elasticsearch</a>, I would simply add that we feel lucky that we have Karel on our team, as it greatly helps us realize our vision around Elasticsearch, and our commitment to our community, specifically, the ruby one.</body></html>

Welcome Karmi

<html><head></head><body>When I first discovered Elasticsearch, in late 2010, I immediately considered it an elaborate hoax.What else can you think when you discover a software project which comes sprinkled with so many buzzwords: a RESTful HTTP interface, working with schema-free documents, talking in JSON through a rich DSL, horizontally scalable and distributed by design, cloud-ready and cloud-friendly? And that software project is a search engine? A hoax, clearly.But then I downloaded the source code, had a look a it... and it seemed quite legit. I became suspicious - it seemed like a little bit too much work for a prank. So I installed said project, and, to my astonishment, it did seem to work the way it was advertised on the tin.At that time, I was working as a lead developer for a social media monitoring application, which was using CouchDB and CouchDB-Lucene as core technologies, so, as you can easily imagine, my neural paths had just exploded like fireworks. Working with schema-free JSON-based documents over HTTP and using full text queries and facets was something I was used to, only this time, it looked way, way better.Being a Ruby shop, we naturally needed a Ruby library for talking to Elasticsearch, and not being satisfied with the offering, I set out to write a <a href="https://github.com/karmi/tire" target="_blank">Ruby gem for Elasticsearch</a>. In spite of working hard to make it usable outside of our pretty specific use case, I never imagined it will get nearly nine hundreds of GitHub watchers and be <a href="http://railscasts.com/episodes/306-elasticsearch-part-1" target="_blank">featured on Railscasts</a> in the process. Or that I will get to solve more then three hundred issues and pull requests, curating other people&apos;s code and making sure the library does not crumble under the weight of different styles, approaches and opinions.In no time, I was writing articles for the Elasticsearch site about <a href="http://www.elasticsearch.org/blog/data-visualization-with-elasticsearch-and-protovis/" target="_blank">creating data visualizations from search facets</a>, <a href="http://www.elasticsearch.org/tutorials/2012/08/22/javascript-web-applications-and-elasticsearch.html" target="_blank">writing self-hosted applications in JavaScript</a>, <a href="http://www.elasticsearch.org/tutorials/2012/03/21/deploying-elasticsearch-with-chef-solo.html" target="_blank">installing Elasticsearch with Chef</a>, and in the process got to write a <a href="https://github.com/karmi/cookbook-elasticsearch" target="_blank">Chef cookbook</a>, Elasticsearch data <a href="https://github.com/karmi/ember-data-elasticsearch" target="_blank">adapter</a> for Ember.js, a monitoring <a href="https://github.com/karmi/elasticsearch-paramedic" target="_blank">tool</a> to check for health problems of the cluster. I <a href="http://www.slideshare.net/karmi/your-data-your-search-elasticsearch-euruko-2011" target="_blank">presented</a> about Elasticsearch at three conferences. In short, my life became fully immersed in this crazy project. During all that time, Shay was an immensely helpful source of information, advice, and inspiration for cranking out features with incredible speed.Today marks a special day on this journey. As a member of the Elasticsearch.com team, I will be able to put more of my time and energy into Elasticsearch and the surrounding ecosystem &#x2014; the Ruby library, provisioning and monitoring tools, research and documentation. My specific goal is to make it even more easy for Rubyists to work with full text search in their applications, to understand it better, and to make most of it.After all, search is the primary means for keeping up with the vast amounts of data; every day, from Google to Mac OS X Spotlight, we rely on search to find information. In fact, the future of our civilization depends on how well we understand the data about ourselves, be it the global Twitter chatter, air pollution metrics, DNA sequences or evidence gathered by the Curiosity rover on Mars. I believe that full text search, and Elasticsearch in particular, will be an important part of this future.</body></html>

the future is elastic

<html><head></head><body>elasticsearch version <code>0.19.9</code> is out. You can download it <a href="/download">here</a>. It contains several bug fixes but also new features, including:
<h2>Snappy Compression</h2>
New experimental support for <code>snappy</code> compression on top of the current <code>lzf</code>. Changing to use it is as simple as setting <code>compress.default.type</code> to <code>snappy</code>. Note, anything already compressed with <code>lzf</code> will still work correctly.
<h2><code>text</code> renamed to <code>match</code></h2>
The <code>text</code> query name gave it great injustice, as it nicely also handles dates and numeric types as well for example. It was renamed to <code>match</code> in order to better reflect what it actually does, with backward support for the <code>text</code> name as well.
On top of that there is now support for <a href="https://github.com/elasticsearch/elasticsearch/issues/2153">multi_match</a> query, allowing to more easily run <code>match</code> queries on several fields without the verbosity of using <code>bool</code> query to wrap it.
<h2>Explain API</h2>
New API called <a href="https://github.com/elasticsearch/elasticsearch/issues/2184">explain</a>, allowing to easily check how and why a specific document matches a query, or even understand why a specific document does not match a specific query.
<h2>Networking Enhacements</h2>
Several networking enhancements geared towards reducing further the amount of buffer copies while doing network operations, and better defaults for network buffer sizes.
</body></html>

0.19.9 Released

<html><head></head><body>
	About 7 years ago, I met with Shay in a small coffee place in Tel Aviv. We talked about Compass and search in general. I only wished I had recorded that conversation. We talked about the future of information retrieval, about big data (although we didn&apos;t use this label yet) and the role that search technology, specifically open source search technology, will have in the future.

	Fast forward a few years and there I was, with years of Lucene and Solr experience behind me, about to set up a company around open source search. I sat down with Steven and together we laid out the blueprints for the next generation search engine. We had a pretty good idea on the direction the information and data management world is heading. And we also knew what it was missing - A simple yet powerful distributed search engine, built to scale from ground up. Sounds pretty simple, doesn&apos;t it :).

	Few months later, Shay released the first version of Elasticsearch. And as time passed, it became clearer to us all that our visions are too much aligned to be ignored. It took some time and effort, but we finally made it happen - today the Elasticsearch company was born.

	It is the beginning of a new era. And not in the traditional definition of search, but in the modern definition of information retrieval. The definition by which data becomes accessible in the form of insightful knowledge, or simply put:
<h3>Making Sense of Your (Big) Data</h3>
	Shay did an amazing job with Elasticsearch. It&apos;s simply mind blowing to realize that he, single handedly, managed to build, support, consult and promote such as amazing piece of software and get it to where it is today. But even kimchyman has his limits. The community is immense. The install base is growing by the day (just the other day Elasticsearch reached the 1 million downloads mark on GitHub). The mailing lists and IRC channels are busier than ever. Companies are running their most critical runtime systems on top of Elasticsearch and demand continuous support. Elasticsearch, the company, no longer the missing piece in the success story of this product. It is here to support the continuation and development of the project, provide the much needed security for the ever growing install-base, and help expand the community through training and education.

	There&apos;s a long journey ahead of us, but an exciting one. And we invite you to join us in this journey and share the excitement.
</body></html>

Exciting Times Ahead

<html><head></head><body>Wow! I&apos;m excited to write this blog today being the day that Open-Source Search moved a huge step forward. I have been working with Lucene for over 8 years now spending an enormous amount of time on adding features, fixing bugs, writing documentation and eventually bringing the project forward together with an awesome gang of committers. I think I can state that Lucene has become one of the most successful Open-Source libraries serving million of users worldwide. Yet, there was always something missing in the picture when you need to go out and build a search system especially when it gets to reliability, fail-over and distributed search. You know those things are hard and have always been out of scope for the Lucene library. A couple of years ago Elasticsearch came into the picture offering an extremely promising idea by an extremely passionate engineer. What can I say, &#x201C;it&apos;s grown up&quot;, here comes Elasticsearch the company!With the rise of NoSQL, Hadoop and being Big-Data everywhere Lucene&apos;s star didn&apos;t shine that bright without offering all these features out of the box and related projects like Solr are slowly catching up with scaling capabilities. But going from an idea to a mature and stable software is a pretty rocky path and it takes time and effort to go. Elasticsearch has gone down that path with all the hard problems in mind from day one. Hey, this is awesome - you know, for search! However, it&apos;s not just Buzz! 4 weeks ago at the Berlin Buzzwords conference you could literally feel how much momentum Elasticsearch gained in the last couple of years. Rooms were packed, engineers talking about success stories, limitations, features and improvements all over the place. This is healthy in a lot of ways and a good indicator that it&apos;s time to take Elasticsearch to the next level. When Shay, Steven, Uri and myself first talked about the idea to join forces and to establish Elasticsearch as a company it was already obvious to me that this is an enormous chance for Open-Source Search in general as well as Lucene being the underlying technology. As a company we heavily rely on the features, maturity and stability of the software we incorporate into Elasticsearch and in turn on the health of the Open Source community. It&apos;s our job and responsibility to maintain it and push it forward. I&apos;m looking forward to a great time working together with great engineers on hard problems making the Elasticsearch user experience as smooth as possible and making Lucene the only choice when you need Search - and trust me you need it!</body></html>

The Rise of Open-source Search

<html><head></head><body>When I set down to write the first lines of code for Elasticsearch, about 3 years ago, I looked at my wife and my 2 months old daughter and knew perfectly what I was getting into. Well, as Miracle Max would say, I mostly did. I knew that its going to be a commitment that will take a big chunk of my life to follow through, I knew that I was building something useful that will make developers life simple, and I knew that there is a need for something like Elasticsearch out there.Obviously, I knew all those things, but not many others did. Its funny, as Tim Robbins found out in the classic Cohen brothers movie, &#x201C;The Hudsucker Proxy&quot;, getting people to see a circle and state &#x201C;You Know, for Kids!&quot;, and making the leap to understand what it can actually be is not simple.Thomas Jefferson said: &#x201C;I am a great believer in Luck, and I find that the harder I work, the more I have of it&quot;. Elasticsearch was and is certainly not a walk in the park. Building a distributed system, that can handle massive amounts of data, and still be usable is no simple task. I do feel lucky though, we live in an age where anyone can open a laptop and set out to change the world (a bit), with extra luck points for having an understanding spouse.More than that, I feel lucky with the community that evolved around Elasticsearch. It caught me by surprise the speed at which Elasticsearch got adopted. I knew that something was missing, I just couldn&apos;t believe how quickly people will realize it as well. I am very proud of what happened around Elasticsearch, the ecosystem that developed around it, and the users actually taking and using it in real systems and in production.But, dodging buses like crazy for the past couple of years has left me exhausted, and the user&apos;s request for having something formal around Elasticsearch, the need to feel safe, has been increasing exponentially. You see, I have seen it happen for 10 years now, it usually starts with &#x201C;lets just have a search box so people can search some content&quot;, and quickly evolves to using it in many other parts of the stack / system, quickly increasing its importance in the application.Luckily (do you notice a trend?), a few months ago, my good friend from the Compass days Uri Boness mentioned that he was doing something around search, and I got to know Steven, who was leading it. Steven and myself immediately hit it off. You see, Steven and myself think very much alike, yet still differently. It&apos;s a rare combination that doesn&apos;t happen frequently, but when it does, exciting things happen. And to top it all, Steven comes with an extensive track record when it comes to Open Source projects. When I learned that Simon Willnauer, one of Lucene rock stars is part of the team as well, the inevitable conclusion was not that hard to make. It Just Felt Right.So, I am extremely happy to announce that we now have an Elasticsearch company, your basic one stop shop for anything to do with Elasticsearch. What does it mean? It means that we can basically move Elasticsearch harder, faster, better, and stronger, while providing all the services you might expect from an Open Source company.On top of that, the Elasticsearch team also includes Nick White (the finance wizard), as the CFO. Chris Male, and Martijn van Groningen, both amazingly talented developers and Lucene committers, and the valuable Elissa Nancarrow to handle, well, basically everything else. With the future holding additional talented people joining the company (and if you are up to it, we are hiring!), it really feels like we are on our way to build something really special.I still vividly remember 10 years ago sitting in a one room apartment in London, with no job, first getting into the search space by writing &#x201C;iCook&quot; for my wife while she was studying to be a Chef at the Cordon Bleu. Its been a long journey to get to this point, yet it feels like it has just begun...<a href="http://www.elasticsearch.com/blog/introducing-elasticsearch-the-company">Steven&apos;s Post</a> 
<a href="http://www.elasticsearch.com/blog/exciting-times-ahead">Uri&apos;s Post</a> 
<a href="http://www.elasticsearch.com/blog/the-rise-of-open-source-search">Simon&apos;s Post</a></body></html>

You Know, for Search! (inc)

<html><head></head><body>Hi everyone,Today is a great day for the Elasticsearch project, all Elasticsearch users, open source in general and not in the least for my colleague, fellow company founder and originator of Elasticsearch - Shay Banon. Today is the day we launch the company behind Elasticsearch! In practice, this means that as of today, users and potential users of Elasticsearch have a definitive source for support, education and guidance with respect to developing, deploying and running Elasticsearch in production environments.Shay and I connected though a mutual acquaintance (company co-founder Uri Boness) and started to share ideas about setting up a company around Elasticsearch. As it turned out, Shay and I shared a vision of creating an innovative company that will really make a difference in the Big Data space, and is focused 100% at creating products that truly meet the needs of users. With my background as co-founder of SpringSource, the company behind the popular Spring Framework, open source was an obvious ingredient of course. It didn&apos;t take us long come to the conclusion that cooperating on serving customers and helping companies make successful use of Elasticsearch made a lot of sense. Only months later we setup a company that is committed to the success of Elasticsearch users and that is committed to open source. We take these commitments very seriously. I am therefore very pleased to state that Elasticsearch will also make contributions to the Apache Lucene project, which is for obvious reasons an open source project that&apos;s very close to our hearts. All the more reason for me to also be extremely excited of having one of the leaders of the Apache Lucene project, Simon Willnauer, as a co-founder of the company and technology leader on my team.Moving forward, Shay will continue to develop Elasticsearch, and will continue to develop a solution that&apos;s exactly what users want and what users need. As of right now, he has a whole team to back him up. During the next few months we will be adding talented Elasticsearch colleagues to the team, so if you&apos;re interested in being part of a great team of technologists in the big data search and analytics space, feel free to drop us a line!I truly look forward to working together with my wonderful team towards the continued success of Elasticsearch. It&apos;s going to be great.Cheers, 
Steven Schuurman, CEO</body></html>

Introducing: Elasticsearch – the Company!

<html><head></head><body>elasticsearch version <code>0.19.8</code> is out. You can download it <a href="/download">here</a>. The release includes bug fixes including integer overflow when reading the new compressed stored format, and a fix for failure to of recovery of peer shards due to failure to detected comrpessor (not common).
<h2>Dangling Indices</h2>
It also includes better handling of dangling indices. Dangling indices happen when a node that has several indices stored locally, joins a cluster and those local indices do not exists in the cluster metadata. This usually does not happen, especially not with proper gateway.recover_after_nodes flag, but still, users can by mistake get into this state.
A new setting <code>gateway.local.auto_import_dangled</code> setting, with possible values of <code>no</code> (never import dangling indices, but also delay the delation of them), <code>yes</code> (import dangling indices), and <code>closed</code> (import dangling indices, but in closed state). The default value is <code>yes</code>.
This also allows to recover &#x201C;other&#x201D; indices into a cluster, by mv&#x2019;ing it into an existing node or a new node, where they will be detected as dangling and be recovered (or imported) into the cluster automatically. Actually, it also has a nice trick of possibly using a &#x201C;new&#x201D; index name based on the name of the index on the file system.
<h2>Wildcard Multi Index Syntax</h2>
Multi index syntax has just become considerably simpler. Current multi index sytax allows to specify several indices to search on using <code>test1,test2,test3</code>, now, it also support wildcards, for example: <code>test*</code>. It also support the ability to &#x201C;add&#x201D; (<code>+</code>) and &#x201C;remove&#x201D; (<code>-</code>), for example: <code>+test*,-test3</code>.
</body></html>

0.19.8 Released

<html><head></head><body>elasticsearch version <code>0.19.7</code> is out. You can download it <a href="/download">here</a>. The release includes mainly bug fixes, fixing the broken debian package, and a bug in the new store compression option (on reads).
</body></html>

0.19.7 Released

<html><head></head><body>elasticsearch version <code>0.19.5</code> is out. You can download it <a href="/download">here</a>. The release includes critical bug fixes in elasticsearch cluster management and peer recovery process, and it is highly recommended for all users.
[Update]: Some users reported a problem when starting elasticsearch with <code>0.19.5</code> due to verification error. This problem has been fixed in <code>0.19.6</code>. 
p. [Update 2]: The <code>deb</code> package seems to be broken in <code>0.19.5</code> and <code>0.19.6</code>, a fix will be issued later. 
p. [Update 3]: <code>0.19.7</code> has been released fixing the debian packaging problem and a bug in the new compression support (on reads).
The release also includes several features, including:
<h2>Store Compression</h2>
elasticsearch has had the ability to compress the _source document for a long time. The problem with it is the fact that small documents don&#x2019;t end up compressing well, as several documents compressed in a single compression &#x201C;block&#x201D; (we use LZF, but Snappy, whihc will be supported also in the future, works the same) will provide a considerable better compression ratio. This version introduces the ability to compress stored fieds using the <code>index.store.compress.stored</code> setting, as well as term vector using the <code>index.store.compress.tv</code> setting.
The settings can be set on the index level, and are dynamic, allowing to change them using the index update settings API. elasticsearch can handle mixed stored / non stored cases. This allows, for example, to enable compression at a later stage in the index lifecycle, and optimize the index to make use of it (generating new segmetns that use compression).
Best compression, comprared to _source level compression, will mainly happen when indexing smaller documents (less than 64k). The price on the other hand is the fact that for each doc returned, a block will need to be decompressed (its fast though) in order to extract the document data.
<h2>Store Throttling</h2>
The way Lucene, the IR library elasticsearh uses under the covers, works is by creating immutable segments (up to deletes) and constantly merging them (the merge policy settings allow to control how those merges happen). The merge proces happen in an asyncronous manner without affecting the indexing / search speed. The problem though, especially on systems with low IO, is that the merge process can be expensive and affect search / index operation simply by the fact that the box is now taxed with more IO happening.
The store module now allows to have throttling configured for merges (or all) either on the node level, or on the index level. The node level throttling will make sure that out of all the shards allocated on that node, the merge process won&#x2019;t pass the specific setting bytes per second. It can be set by setting <code>indices.store.throttle.type</code> to <code>merge</code>, and setting <code>indices.store.throttle.max_bytes_per_sec</code> to something like <code>5mb</code>. The node level settings can be changed dynamically using the cluster update settings API.
If specific index level configuration is needed, regardless of the node level settings, it can be set as well using the <code>index.store.throttle.type</code>, and <code>index.store.throttle.max_bytes_per_sec</code>. The default value for the type is <code>node</code>, meaning it will throttle based on the node level settings and participate in the global throttling happening. Both settings can be set using the index udpate settings API dynamically.
</body></html>

0.19.5 Released

<html><head></head><body>elasticsearch version <code>0.19.4</code> is out. You can download it <a href="/download">here</a>. The release includes serveral bug fixes, mainly around percolator, <code>_default_</code> mapping type, and file based config index templates.
It also includes an enahcement allowing to configure the <code>query_string</code> query to be <code>lenient</code>, not failing when wrong format of a value is provided (for example, a text to a numeric field), as well as the ability to search &#x201C;within&#x201D; objects using wildcards in the <code>query_string</code> syntax (it was already provided if it was used in the <code>fields</code> section), for example: <code>city.\*:something</code>.
</body></html>

0.19.4 Released

<html><head></head><body>elasticsearch version <code>0.19.3</code> is out. You can download it <a href="/download">here</a>. The release includes an upgraded Lucene 3.6 version, and several bug fixes, including better handling of network splits with node client nodes and with minimum master nodes.
</body></html>

0.19.3 Released

<html><head></head><body>elasticsearch version <code>0.19.2</code> is out. You can download it <a href="/download">here</a>. It is a bug fix release fixing several bugs including a bug not being able to update index level settings on a live index using the index update settings API.
</body></html>

0.19.2 Released

<html><head></head><body>elasticsearch version <code>0.19.1</code> is out. You can download it <a href="/download">here</a>. It is a bug fix release fixing several bugs including a major bug where a thread can get into CPU spin when its relocated and a search/stats operation is executed against it.
The release also includes an update to the `read_only` option on an index. The option caused both write and metadata operations to be disabled for the relevant index. Now, in 0.19.1, specific read, write and metadata blocks can be set. See more <a href="https://github.com/elasticsearch/elasticsearch/issues/1771">here</a>.
</body></html>

0.19.1 Released

<html><head></head><body>elasticsearch version <code>0.19.0</code> is out. You can download it <a href="/download">here</a>. It is the final 0.19.0 release (after 3 release candidates). 0.19 major features have been outlined in the previous blog posts for each release candidate release, but, it also marks an important bug fix and stability release over 0.18, upgrade is highly encouraged to all elasticsearch users.
</body></html>

0.19.0 Released

<html><head></head><body>elasticsearch version <code>0.19.0.RC3</code> is out. You can download it <a href="/download">here</a>. Its another bug fix released and (hopefully) the final release of 0.19.0 release candidates.
It also includes two nice features. The first is the ability to use &#x201C;date math&#x201D; on <code>date</code> types (especially useful in <code>range</code> queries/filters). See more info in the <a href="https://github.com/elasticsearch/elasticsearch/issues/1708">issue</a> (and documented under the mapping date type).
The second feature is a new API called &#x201C;Multi Search&#x201D; (or <code>msearch</code>). It allows to construct a single request holding several search requests, executing them in the cluster (in parallel) and returning the results. More info on the format and how to use it can be found in the <a href="https://github.com/elasticsearch/elasticsearch/issues/1722">issue</a> (and documented under the guide API section).
</body></html>

0.19.0.RC3 Released

<html><head></head><body>elasticsearch version <code>0.19.0.RC2</code> is out. You can download it <a href="/download">here</a>. Its a bug fix release mainly fixing a major bug in how buffers are read and reused over the network (the optimization done in 0.19.0.RC1 was not complete).
It also include two nice features, the first is the ability to disallow a shard and its replica to be allocated on the same machine when running more than one instance on said machine. It can be enabled by setting <code>cluster.routing.allocation.same_shard.host</code> to <code>true</code>.
The second is an effort to better log GC pause times (without needing to turn on GC logging), with GC type thresholds and logging thresholds (comes with built in default values).
</body></html>

0.19.0.RC2 Released

<html><head></head><body>elasticsearch version <code>0.19.0.RC1</code> is out. You can download it <a href="/download">here</a>. It is a major release candidate release of elasticsearch.
This is the first time we release a &#x201C;release candidate&#x201D; for elasticsearch. The aim is to get users to try it out, fix any problems found with it, and release a final <code>0.19.0</code> release as soon as possible.
Major features include improved indexing performance, more control over shard allocation, more statistics and simpler API for stats, an update API that accepts a script to perform an update on a document, better state storage when using local gateway, and many bug fixes.
<h2>Upgrading</h2>
This release requires a full cluster restart in order to upgrade to the new version (including issuing a flush across all indices before the restart). If you are using the (default) local gateway, it will automatically be upgraded to a new and improved state format (with backups of the original state).
The structure of the project has also changed, moving all the plugins to their own repos under <a href="http://github.com/elasticsearch/">github</a>. Installing a plugin now requires specifying the location and version to install it from.
</body></html>

0.19.0.RC1 Released

<html><head></head><body>elasticsearch version <code>0.18.7</code> is out. You can download it <a href="/download">here</a>. It includes bug fixes including major bug fix in state storage with shared gateways (s3 for example), and support for nested documents in delete by query.
</body></html>

Elasticsearch Platform

ELK Stack

Elastic Cloud

Observability

Security

Search

By industry

By solution

Customer spotlight

Developers

Connect

Learn

Help

See what's happening at Elastic

<

2012

<

december

0.20.2 Released

New Download Service

0.20.0 Released

0.19.12 Released

Follow us

About us

Join us

Press

Partners

Trust & Security

Investor relations

EXCELLENCE AWARDS