Explaining Ark Part 4: Fixing Majority Write...

Posted on by Zardosht Kasheff

This is the fourth post in a series of posts that explains Ark, a consensus algorithm we’ve developed for TokuMX and MongoDB to fix known issues in elections and failover. The tech report… we released describes the algorithm in

Leave a comment

An Updated Description of Clustering Keys for...

Posted on by Rich Prohaska

Covering indexes can result in orders of magnitude performance improvements for queries. Bradley’s presentation on covering indexes… describes what a covering index is, how it can effect performance, and why it works. However, the definition of a covering index can

Leave a comment

New TokuMX Meetup Groups Formed

Posted on by Dave Rosenlund

Tokutek is pleased to announce the formation of two new TokuMX Bay Area Meetup Groups, as well as each group’s inaugural meet up event. These groups are for anyone who’s interested in delivering high-performance big data applications built with the…

Leave a comment

Explaining Ark Part 3: Why Data May Be Lost on a...

Posted on by Zardosht Kasheff

This is the third post in a series of posts that explains Ark, a consensus algorithm we’ve developed for TokuMX and MongoDB to fix known issues in elections and failover. The tech report… we released last week describes the

Leave a comment

Explaining Ark Part 2: How Elections and Failover...

Posted on by Zardosht Kasheff

This is the second post in a series of posts that explains Ark, a consensus algorithm we’ve developed for TokuMX and MongoDB to fix known issues in elections and failover. The tech report… we released last week describes the algorithm

Leave a comment

Explaining Ark Part 1: The Basics

Posted on by Zardosht Kasheff

Last week, we introduced Ark, a consensus algorithm similar to Raft and Paxos we’ve developed for TokuMX and MongoDB. The purpose of Ark is to fix known issues in elections and failover. While the tech report detailing Ark… explains everything

Leave a comment

Introducing Ark: A Consensus Algorithm For TokuMX...

Posted on by Zardosht Kasheff

Most of the time, our blog posts explain what’s great about the MongoDB improvements we’ve already shipped in TokuMX…. Sometimes, though, it’s fun to talk about what’s coming soon, especially when user feedback would really help get the feature

4 Comments

TokuMX Compression & Performance Innovation...

Posted on by Dave Rosenlund

Today, at MongoDB World 2014 in New York City, we announced the winners of our inaugural TokuMX Compression & Performance Innovation Award. We selected two winners from amongst the nominees – Brightcove and Crowdtap…. The announcement was made at

Leave a comment

TokuMX @ MongoDB World 2014

Posted on by Dave Rosenlund

On Monday, some of the Boston-area Tokutek team will board New York bound trains to join our Union Square colleagues at MongoDB World 2014…. We couldn’t be more excited.
You would be too. TokuMX, our high-performance distribution of MongoDB,

Leave a comment

Use TokuMX Partitioned Collections in Place of TTL...

Posted on by Zardosht Kasheff

Take the following scenario. You have a time-series data… application for which you would like to store a rolling period of data. For example, you may want to maintain the last six months of traffic logs for a website, in

Leave a comment