Home |  MySQL Buzz |  FAQ |  Feeds |  Submit your blog feed |  Feedback |  Archive |  Aggregate feed RSS 2.0 English Deutsch Español Français Italiano 日本語 Русский Português 中文
Previous 30 Newer Entries Showing entries 91 to 120 of 248 Next 30 Older Entries

Displaying posts with tag: TokuDB (reset)

Fast Updates with TokuDB
+4 Vote Up -0Vote Down

With TokuDB v6.6 out now, I’m excited to present one of my favorite enhancements: fast updates with TokuDB. Update intensive applications can have their throughput limited by the random read capacity of the storage system. The cause of the throughput limit is the read-modify-write algorithm that MySQL uses when processing update statements. MySQL reads a row from the storage engine, applies the updates to it, and then writes the new row to the storage engine. To address this throughput limit, TokuDB uses a different update algorithm that simply encodes the update expressions of the SQL statement into tiny programs that are stored in an update Fractal Tree® message. This update message is

  [Read more...]
Concurrency Improvements in TokuDB v6.6 (Part 2)
+0 Vote Up -0Vote Down

In Part 1, we showed performance results of some of the work that’s gone in to TokuDB v6.6. In this post, we’ll take a closer look at how this happened, on the engineering side, and how to think about the performance characteristics in the new version.

Background

It’s easiest to think about our concurrency changes in terms of a Fractal Tree® index that has nodes like a B-tree index, and buffers on each node that batch changes for the subtree rooted at that node. We have materials that describe this available here, but we can proceed just knowing that:

  • To inject data into the tree, you need to store a message in a buffer at the root of the tree. These
  •   [Read more...]
    Concurrency Improvements in TokuDB v6.6 (Part 1)
    +2 Vote Up -0Vote Down

    With TokuDB v6.6 out now, I’m excited to present one of my favorite enhancements: concurrency within a single index. Previously, while there could be many SQL transactions in-flight at any given moment, operations inside a single index were fairly serialized. We’ve been working on concurrency for a few versions, and things have been getting a lot better over time. Today I’ll talk about what to expect from v6.6. Next time, we’ll see why.

    Summary of Results

    Running multiple iiBench clients on a single MySQL instance, we see a big improvement in the cumulative insertion speed at all concurrency levels. We see a gain of 33.9% in single-threaded performance and 51.8% at

      [Read more...]
    Tracking 5.3 Billion Mutations: Using MySQL for Genomic Big Data
    +1 Vote Up -0Vote Down

    University of Montreal Tracks Genomic Data With Tokutek’s TokuDB.

    Faster insertion rates, improved scalability and agility support lab’s fast growing research database as it grows from 100s of GBs to 1 TB and beyond.

    Issue addressed: MySQL database used for genomic research must be able to quickly ingest huge amounts of incoming data – hundreds of thousands of records every day. It also must be able to retrieve data quickly in response to a diverse set of research requests.

    Enabling the Hunt for New Cures for Diseases by Seamlessly Processing Billions of Mutations  [Read more...]

    The Results Are In!
    +1 Vote Up -0Vote Down

    We wanted to take a moment to say thanks to all of our customers and to the wider MySQL and MariaDB community. Today we announced a doubling of our customer base for the year ending December 31, 2012. Significant milestones over the last year included new technology and service partnerships, several awards, rapid hiring, as well as three upgrades to TokuDB®. We even dabbled in some MongoDB benchmarks. And to fuel continued growth in 2013, we secured additional venture capital funding last November.

    Did You Hear? NASA Uses TokuDB for Big Data with MySQL!

    To read the full press release and learn more,

      [Read more...]
    Webinar: Introduction to TokuDB v6.6
    +2 Vote Up -0Vote Down

    TokuDB® is a proven solution that scales MySQL® and MariaDB® from GBs to TBs with unmatched insert and query speed, compression, replication performance and online schema flexibility. Tokutek’s recently launched TokuDB v6.6 delivers all of these features and more, with additional improvements in multi-client, fast SQL updates, and in-memory performance.

    Date: January 15th
    Time: 2 PM EST / 11 AM PST
    REGISTER TODAY

    Topics will include:

    • Performance – With a 10x or more improvement in insertions and indexing, TokuDB delivers faster, more complex ad hoc queries in live production systems without rewriting or tuning applications. Offering high performance even when tables are too large for memory, TokuDB scales MySQL and MariaDB


      [Read more...]
    Announcing TokuDB v6.6: Performance Improvements
    +1 Vote Up -0Vote Down

    We are excited to announce TokuDB® v6.6, the latest version of Tokutek’s flagship storage engine for MySQL and MariaDB.

    This version offers three types of performance improvements: in-memory, multi-client and fast updates.

    Although TokuDB is optimized for large tables, which are larger than memory, many workloads consist of a mix of large and small tables. TokuDB v6.6 offers improvements on in-memory performance, with a more than 100% improvement on Sysbench at many concurrency levels and more than 200% improvement on TPC-C at many concurrency levels. Details to follow.

    We have also made improvements in multi-threaded performance. For example, single threaded trickle loads have always been fast in TokuDB. But now multi-threaded trickle loads are even faster. An

      [Read more...]
    Marinating in 2013
    +0 Vote Up -0Vote Down

    What a flashback this week. Staring at a text terminal trying to establish a connection with a remote server, I began to fret whether I would get my homework assignment done on time. My mind raced back to college nights years ago in the Fishbowl, hunched over an Athena workstation. Would this be another late night fueled by Jolt cola in order to get my problem set done?

    Thankfully, no!

    Embarking on my first software class in quite a while was relatively painless, and I have Sheeri Cabral and her detailed guidance to thank. This week I started the

      [Read more...]
    Move over Marcia: Top Ten for 2012
    +0 Vote Up -0Vote Down

    Well, it’s that time of the year again for top ten lists. There have been many versions showing up on the web the last few days, including Time Magazine’s “Top 10 Everything of 2012″ list, with 55 wide ranging lists!

    Last year we started using Google Analytics to see what content for blogs was most popular on Tokutek.com and generated a 2011 top ten list, ending up with a few surprises.  This year saw spikes in some interesting areas as well, including flash performance, NASA and Big Data, and MongoDB.

    Without further adieu, here is the top ten list for 2012:

    10. Announcing TokuDB v6.1

      [Read more...]
    Packing for the Holidays
    +2 Vote Up -0Vote Down

    Every time I visit my family for the holidays, as the date approaches, I find myself filled with dread. It’s nothing sinister, my family’s great, and the season is nice. The reason is simple:

    I hate packing.

    In fact, I hate both kinds of packing: trip packing, and bit packing. Let me tell you a story about bit packing.

    Bit packing

    If you’ve ever browsed around the available type attributes in GCC, you may have noticed the entry for “packed”. It seems straightforward enough, and if you’re trying to cram a lot of data in a system (like we do), it can be pretty attractive.

    There are plenty

      [Read more...]
    Yule Blog
    +1 Vote Up -0Vote Down

    Tired of the same old holiday tunes year after year? Join us as we Rock around the Fractal Tree here at Tokutek.

    We’ve fiddled with the chorus; won’t you join us?

    Rockin’ around the Fractal Tree
    (sung to the tune of Rockin’ around the Christmas Tree)

    Rockin’ around the Fractal Tree
    As the bits begin to hop
    Leaves are full on trees that are not B
    Our indexes cannot be stopped

    Rockin’ around the Fractal Tree
    Let the MySQL spirit ring
    Time to compress all into an SSD
    And watch out for the Merciless Ming

    You will get a sensational feeling when you hear voices singing
    “No, not a fable; online changes done with alter table!”
    Rockin’ around the Fractal Tree










      [Read more...]
    Fractal Tree Indexing Overview
    +0 Vote Up -1Vote Down

    We get a lot of questions about how Fractal Tree indexes work. It’s a write-optimized index with fast queries, but which write-optimized indexing structure is it?

    In this ~15 minute video (which uses these slides), I give a quick overview of how they work and what they are good for.

    Webinar: Best Practices for a Successful TokuDB Evaluation
    +0 Vote Up -0Vote Down

    In this webinar we will show step by step how to install, configure, and test TokuDB for a typical performance evaluation. We’ll also be flagging potential pitfalls that can ruin the eval results. It will describe the differences between installing from scratch and replacing an existing MySQL / MariaDB installation. It will also review the most common issues that may arise when running TokuDB binaries.

    Date: December 11th
    Time: 2 PM EST / 11 AM PST
    REGISTER TODAY

    Topics will include:

    • Memory allocation
    • Initial data load
    • Indexing, including clustering indexes
    • compression algorithms

    We look forward to having you join



      [Read more...]
    Scalable Databases for Startups
    +0 Vote Up -0Vote Down

    One of the great things about the MassTLC unConference is the spontaneity of the ideas. In the morning I ran into an old colleague whose startup was looking at switching databases and struggling with the options. Hence, “Scalable Databases for Startups” seemed like a great topic, so I proposed it, and then was off and running full steam after lunch.

    The session brought in a wide variety of firms. While there were several vendors there – Basho, Calpont, InterSystems,

      [Read more...]
    Small Data
    +0 Vote Up -0Vote Down

    There is obviously much being written these days about Big Data. While the term has many different meanings to many different folks, our MySQL and MariaDB customers tend to find their data to be uncomfortably big when the tables become too large for memory. In this case, more storage has to be acquired, performance starts to lag, and making changes to the schema becomes a challenge.

    TokuDB addresses these issues for big MySQL instances by delivering high compression rates, faster insertion and query performance, and agile

      [Read more...]
    Join us at ad:tech this week
    +0 Vote Up -0Vote Down

    It certainly has been a tough week on the East Coast. In the Boston area, where I am located, we got grazed by Sandy – power outages, trees down, schools closed, and Halloween delayed. This though, pales in comparison to what our NY office had to endure. One of our co-founders for example is one of the first few to get his power back at his home on Long Island just yesterday. He considers himself lucky – one of the several pine trees that came down on his property was just a few degrees off from destroying his house (it ended up taking down the gutter instead). Needless to say, it’s going to be a long recovery for NY and NJ. If you are looking to help, see here.

    Of course, NYC is known for its resiliency. The first post-Sandy event at Manhattan’s Javitis Conference Center, 

      [Read more...]
    Presenting at tomorrow’s NoVA MySQL Meetup (DC Area)
    +1 Vote Up -0Vote Down

    At tomorrow’s NoVA MySQL October Meetup, I will give a talk: “Fractal Tree Indexes – Theoretical Overview and Customer Use Cases.” The meetup is 7 pm Tuesday, October 23, 2012, and will be held at AOL Campus HQ in Dulles VA.

    Most databases employ B-trees to achieve a good tradeoff between the ability to update data quickly and to search it quickly. It turns out that B-trees are far from the optimum in this tradeoff space. This led to the development at MIT, Rutgers and Stony Brook of Fractal Tree® indexes. Fractal Tree indexes improve MySQL® scalability and query performance by allowing greater insertion rates, supporting rich indexing and offering

      [Read more...]
    My Talk Next Week at HighLoad++
    +2 Vote Up -0Vote Down

    Next week I’ll be visiting Moscow to talk at Highload++. The conference will take place during Monday 22nd and Tuesday 23rd at the Radisson hotel. I will be giving my personal version of an indexing talk that my colleagues have given in meetups and conferences in the US.

    Highload++ conference is targeted to address the issues of complex high traffic web properties. Most of these sites depend on databases to deliver their content, record the traffic and report the application activities in real time. As I learned early in my career at MySQL, the database schema and in particular the indexing strategy, are critical to achieve the highest possible performance out of the

      [Read more...]
    Report on XLDB Tutorial on Data Structures and Algorithms
    +0 Vote Up -0Vote Down

    Bradley and I (Michael) gave the tutorial on Data Structures and Algorithms for Big Databases at the 6th XLDB Conference last month.

    The tutorial was organized as follows:

    • Module 0: Tutorial overview and introductions. We describe an observed (but not necessary) tradeoff in ingestion, querying, and freshness in traditional database.
    • Module 1: I/O model and cache-oblivious analysis.
    • Module 2: Write-optimized data structures. We give the optimal trade-off between inserts
      [Read more...]
    First of New NSF Big Data Grants Go to Tokutek Founders
    +0 Vote Up -0Vote Down

    The core technology behind Tokutek is based on the academic research by our founders: Michael Bender, Bradley Kuszmaul and Martin Farach-Colton.  They are all still in academia, in addition to their work at Tokutek.

    Back in March, the White House kicked off a new Initiative for Big Data.  Last week, the National Science Foundation announced the first interagency grants for this.  Eight awards were given, and our own Michael Bender and Martin Farach-Colton, along with Robert Johnson of Stony Brook University, received one of

      [Read more...]
    ODBMS Interview: Scaling MySQL and MariaDB to TBs
    +2 Vote Up -0Vote Down

    Recently, our CTO, Martín Farach-Colton had a chance to talk about scaling MySQL and MariaDB with Roberto Zicari of ODBMS.

    In the article, Martin states “While I believe that one size fits most, claims that RDBMS can no longer keep up with modern workloads come in from all directions. When people talk about performance of databases on large systems, the root cause of their concerns is often the performance of the underlying B-tree index.” He also notes how “Fractal Tree Indexes put you on a higher-performing tradeoff curve. Query-optimal write-optimized indexing is all about making general-purpose databases faster. For some of our customers’ workloads,

      [Read more...]
    Forbes: “Tokutek Makes Big Data Dance”
    +1 Vote Up -0Vote Down

    Recently, our CEO, John Partridge had a chance to talk about novel database technologies for “Big Data” with Peter Cohan of Forbes.

    According to the article, “Fractal Tree indexing is helping organizations analyze big data more efficiently due to its ability to improve database efficiency thanks to faster ‘database insertion speed, quicker input/output performance, operational agility, and data compression.’” As a start-up based on “the  first algorithm-based breakthrough in the database world in 40 years,” Toktuetek is following in the footsteps of firms such as Google and RSA, which also relied on novel algortithm advances as core to their technology.

    To read the full article, and

      [Read more...]
    Webinar: Introduction to TokuDB v6.5
    +1 Vote Up -0Vote Down

    TokuDB® is a proven solution that scales MySQL® and MariaDB® from GBs to TBs with unmatched insert and query speed, compression, replication performance and online schema flexibility. Tokutek’s recently launched TokuDB v6.5 delivers all of these features and more, not just for HDDs, but also for flash memory.

    Originally Aired: October 10th
    AVAILABLE ON DEMAND

    TokuDB v6.5:

    • Stores 10x More Data – TokuDB delivers 10x compression without any performance degradation. Users can therefore take advantage of much greater amounts of available space without paying more for additional storage.
    • Delivers High Insertion Speed – TokuDB

      [Read more...]
    MySQL Schema Agility on SSDs
    +1 Vote Up -0Vote Down

    TokuDB v6.5 adds the ability to expand certain column types without downtime.  Users can now enlarge char, varchar, varbinary, and integer columns with no interruption to insert/update/delete statements on the altered table.  Prior to this feature, enlarging one of these column types required a full table rebuild.  InnoDB blocks all insert/update/delete operations to a table during column expansion as it rebuilds the table and all indexes.

    We recently added SSDs to our existing servers (HP DL380 servers with internal RAID arrays).  I opted for LSI controllers (LSI 9285-8e and LSI 9280-4i4e), Samsung 830 SSDs, and a Sans Digital AccuRaid AS108X enclosure.  I was unable to locate much in the way of optimizing Linux for SSD/RAID performance and plan on blogging my findings in the future.  However, the out of the box performance is seriously fast.

    I was curious to see how

      [Read more...]
    Three Ways that Fractal Tree Indexes Improve SSD for MySQL
    +1 Vote Up -0Vote Down

    Since Fractal Tree indexes turn random writes into sequential writes, it’s easy to see why they offer a big advantage for maintaining indexes on rotating disks. It turns out that that Fractal Tree indexing also offers signficant advantages on SSD. Here are three ways that Fractal Trees improve your life if you use SSDs.

    Advantage 1: Index maintenence performance.

    The results below show the insertion of 1 billion rows into a table while maintaining three multicolumn secondary indexes. At the end of the test, TokuDB’s insertion rate remained at 14,532 inserts/second whereas InnoDB had dropped to 1,607 inserts/second. That’s a difference of over 9x.

    Platform: Centos 5.6; 2x Xeon L5520; 72GB RAM; LSI  [Read more...]
    Announcing TokuDB v6.5: Optimized for Flash
    +2 Vote Up -0Vote Down

    We are excited to announce TokuDB® v6.5, the latest version of Tokutek’s flagship storage engine for MySQL and MariaDB.

    This version offers optimization for Flash as well as more hot schema change operations for improved agility.

    We’ll be posting more details about the new features and performance, so here’s an overview of what’s in store.

    Flash TokuDB v6.5 continues the great Toku-tradition of fast insertions. On flash drives, we show an order-of-magnitude (9x) faster insertion rate than InnoDB. TokuDB’s standard compression works just as well on flash and helps you get the most out of your  [Read more...]
    Tokutek CEO Named Innovation All Star
    +1 Vote Up -0Vote Down

    On Friday our CEO John Partridge was named a “Tech Luminary” in the 17th annual Innovation All Stars award, which is given jointly by Mass High Tech (MHT) and the Boston Business Journal (BBJ). As noted in MHT by the editor, Chris McIntosh, the Luminary designations “reflect deep accomplishment in various technology-related industries.”

    For more than 20 years, Tokutek CEO John Partridge has worked with startups in both the Boston area and in Silicon Valley. He joined Tokutek from StreamBase Systems which John co-founded with

      [Read more...]
    Strange Loop Talk on Indexing
    +0 Vote Up -0Vote Down

    At next week’s Strange Loop conference, I will give a talk on “Understanding Indexing”. The session is 10 am Monday, September 24th, and will be held in the Midland States Room.

    Application performance often depends on how fast a query can respond and query performance almost always depends on good indexing. So one of the quickest and least expensive ways to increase application performance is to optimize the indexes. This talk presents three simple and effective rules on how to construct indexes around queries that result in good performance.

    This is a general discussion applicable

      [Read more...]
    XLDB Tutorial on Data Structures and Algorithms
    +0 Vote Up -0Vote Down

    Next week Michael and I (Bradley) will be travelling to Silicon Valley to present a tutorial on Data Structures and Algorithms for Big Databases at the 6th XLDB Conference.

    The tutorial, which is 4 hours on Monday afternoon, aims to cover the following topics (but it’s looking like we’ll have to drop several items for lack of time.)

    This tutorial will explore data structures and algorithms for big databases. The topics include:

    • Data structures including B-trees, Log Structured Merge Trees, and Streaming B-trees.
    • Approximate Query Membership data structures including Bloom filters and cascade filters.
    • Algorithms for join including hash joins and
      [Read more...]
    268x Query Performance Increase for MongoDB with Fractal Tree Indexes, SAY WHAT?
    +1 Vote Up -0Vote Down

    Last week I wrote about our 10x insertion performance increase with MongoDB. We’ve continued our experimental integration of Fractal Tree® Indexes into MongoDB, adding support for clustered indexes.  A clustered index stores all non-index fields as the “value” portion of the index, as opposed to a standard MongoDB index that stores a pointer to the document data.  The benefit is that indexed lookups can immediately return any requested values instead of needing to do an additional lookup (and potential disk IOs) for the requested fields.

    To create a clustered index you just need to add

      [Read more...]
    Previous 30 Newer Entries Showing entries 91 to 120 of 248 Next 30 Older Entries

    Planet MySQL © 1995, 2014, Oracle Corporation and/or its affiliates   Legal Policies | Your Privacy Rights | Terms of Use

    Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.