Twitter Claims Searches 3x Faster With “Blender”

  • Share
  • Read Later

A new homepage isn’t the only change Twitter’s recently undergone. As the Twitter Engineering blog reports, last week saw the launch of Blender, a new server that reportedly makes its search function three times faster than before.

The new server doesn’t just speed up what the blog calls “one of the most heavily-trafficked search engines in the world,” it distributes the workload.

According to Twitter’s engineers, the introduction of Blender means that Twitter can now handle a higher number of requests than before, ultimately helping reduce costs by allowing Twitter to invest in fewer servers as demand continues to grow.

Want to know how Blender does what it does? Here’s a technobabble sample:

In the sample workflow above, we have 6 services {s1, s2, s3, s4, s5, s6} with dependencies between them. The directed edge from s3 to s1 means that s3 must be called before calling s1 because s1 needs the results from s3. Given such a workflow, the Blender framework performs a topological sort on the DAG to determine the total ordering of services, which is the order in which they must be called. The execution order of the above workflow would be {(s3, s4), (s1, s5, s6), (s2)}. This means s3 and s4 can be called in parallel in the first batch, and once their responses are returned, s1, s5, and s6 can be called in parallel in the next batch, before finally calling s2.

Hope you caught all that, and while it sounds impressive, the blog avoids the more important question: Whether Blender will be able to prevent more involuntary changes between New Twitter and Old Twitter, as happened for an hour or so last night (April 5).

More on

Take A Look At Twitter’s Handsome New Homepage

New Twitter App Puts Your Followers on the Chopping Block

How To Tweet Like Barack Obama (Or Anyone Else, For That Matter)