BigQuery Big Data Analytics Now Available to all Businesses; New V8 Speed Improved by 25%; Introducing Ceres Solver Nonlinear Least Squares Solver

Back in November last year, Google launched "BigQuery" in a limited preview to help businesses and developers start using it for real-time Big Data analytics in the cloud. Today, the comapny is making BigQuery publicly available, to bring Big Data analytics to all businesses via the cloud. "BigQuery is accessible via a simple UI or […]

Back in November last year, Google launched "BigQuery" in a limited preview to help businesses and developers start using it for real-time Big Data analytics in the cloud. Today, the comapny is making BigQuery publicly available, to bring Big Data analytics to all businesses via the cloud.

"BigQuery is accessible via a simple UI or REST interface. It lets you take advantage of Google's massive compute power, store as much data as needed and pay only for what you use. Your data is protected with multiple layers of security, replicated across multiple data centers and can be easily exported," Ju-Kay Kwek, Product Manager, BigQuery, explains.

Developers and businesses can sign up for BigQuery online and query up to 100 GB of data per month for free.

BigQuery Big Data Analytics Web Service

SPDY is a replacement for HTTP, designed to speed up transfers of web pages, by eliminating much of the overhead associated with HTTP. SPDY includes several features that should improve web page download speeds on mobile networks, including:

  1. Header compression, which eliminates redundant data for HTTP headers;
  2. Out-of-order request processing, avoiding head-of-line blocking for HTTP responses;
  3. Use of a single TCP connection for multiple requests, eliminating overheads for TCP connection establishment (which can be high on mobile networks).

Google recently gone through a test to know what the performance of SPDY would be compared to HTTP for popular websites, using a real phone (a Samsung Galaxy Nexus running Android), a modern, SPDY-enabled browser (Chrome for Android), and a variety of pages from real websites (77 pages across 31 popular domains).

"The net result is that using SPDY results in a mean page load time improvement of 23% across these sites, compared to HTTP. This is equivalent to a speedup of 1.3x for SPDY over HTTP. Much more work can be done to improve SPDY performance on 3G and 4G cellular networks, but this is a promising start," posted Matt Welsh, Ben Greenstein, and Michael Piatek.

The following graph shows the page load time for HTTP and SPDY, in milliseconds, across the 77 pages that were measured. As the graph shows, in all but one case, SPDY reduces load times, sometimes by as much as 50%.

SPDY mobile networks performanceGoogle Chrome in the recent most dev and beta channel releases, using a new release of "V8 ," that brings a new algorithm based on counters to decide which functions to optimize. This greatly increases performance for small JavaScript programs. "For example, on the SunSpider benchmark, which focuses on extremely short-running tests, V8's speed improved by about 25%," explains Jakob Kummerow, Software Engineer.

New V8 algorithm

"The new version of V8 makes earlier and more repeatable optimization decisions by analyzing the running program in more detail. It uses counters to keep track of how often JavaScript functions are called and loops are executed in a program, approximating the time spent inside each function. That way V8 is able to quickly gather fine-grained information about performance bottlenecks in a JavaScript program, and to make sure that the optimizing compiler's efforts are spent on those functions that deserve it most," explained Kummerow.

The new algorithm is contained in the current beta channel releases.

Finally, Google today released a non-linear least squares problems solver the comapny use at Google called "Ceres Solver," is a portable C++ library that allows for modeling and solving large complex nonlinear least squares problems.

"We use Ceres Solver at Google to estimate the pose of Street View cars, aircrafts, and satellites; to build 3D models for PhotoTours; to estimate satellite image sensor characteristics, and more," Google posted.

Ceres Solver notable features include:

  • A simple, expressive API
  • Automatic differentiation
  • Robust loss functions
  • Local parameterizations
  • A threaded Jacobian evaluators and linear solvers
  • Dense QR factorization (using Eigen) for small problems
  • Sparse Cholesky factorization (using SuiteSparse) for large sparse problems
  • Specialized solvers for problems in 3D computer vision
  • A liberal license (New BSD)
  • Scales from servers to cell phones.

You can download the code here.