Speed, especially in NASCAR and IT events, makes news. On Nov. 10, server-maker IBM, high-performance networking provider Mellanox and China-based Tencent Cloud made their own IT speed news when a cloud solution they put together was named winner of the Sort Benchmark annual global computing competition for 2016.
Tencent Cloud and the Big Data Collaboration team spent less than 99 seconds (98.8 seconds) finishing sorting 100 terabytes of data in the Indy GraySort category, improving upon 2015’s best result—329 seconds—by a factor of 3.3.
Each year, global companies and academic institutions participate in the contest, conducted by non-profit Sortbenchmark.org, to evaluate the capability of their software and hardware system architectures, as well as their research results.
Tencent Cloud participated in two of the four competition categories, GraySort and MinuteSort, to benchmark the architecture capabilities and computing efficiencies of large distributed computing systems and real time data analytics. The GraySort and MinuteSort benchmarks are each measured in two subcategories: Indy (special purpose sort) and Daytona (general purpose sort).
Dominant in Each Category
Tencent Cloud Data Intelligence—the distributed computing platform of Tencent Cloud—won each category in both the GraySort and MinuteSort benchmarks, establishing four new records with its performance, outperforming 2015’s best speeds by two to five times.
“In the future, the ability to manage big data will be the foundation of successful internet businesses,” said Zeus Jiang, Vice President of Tencent Cloud and General Manager of Tencent’s Data Platform Department. “Tencent Cloud can provide precise high-performance computing to enterprises using minimal time and resources. We will continue to improve the back-end technology for our cloud service by optimizing architecture, software and hardware to help global enterprises solve complex business challenges by leveraging hyper-scale computing platforms.”
As a non-profit organization, Sortbenchmark.org uses the basic sort computing method to evaluate distributed system software architectural capabilities such as storage, task scheduling, node communication and data exchange. It also evaluates hardware architectural capabilities, including server selection and match, network architecture optimization, and the ability to build Internet Data Centers.
Solution Used IBM POWER Servers, Mellanox 100Gb Connectivity
The ultimate test is on the comprehensive capabilities of large-scale computing of a company or an organization. In this sort contest, Tencent Cloud used 512 IBM OpenPOWER servers and Mellanox’ 100Gb interconnect technology, improving the performance of Tencent Cloud big data products with the infrastructure.
“Tencent is helping IBM and our OpenPOWER partners push performance boundaries for a cognitive era defined by big data and advanced analytics,” said Tom Rosamilia, Senior Vice President of IBM Systems. “The computing record achieved by Tencent Cloud on OpenPOWER is an important milestone for the industry.”
Real time-analytics and big data environments are extremely demanding and the network is critical in linking together the extra high performance IBM POWER based servers and Tencent Cloud’s massive amounts of data, said Amir Prescher, SeniorVice President, Business Development, at Mellanox Technologies.
“Tencent Cloud developed an optimized hardware/software platform to achieve new computing records, showing that Mellanox’s 100Gb/s Ethernet technology can deliver total infrastructure efficiency and improves application performance, making them ideal for Big Data applications,” Prescher said.
2016 Sort Benchmark Contest Results (improvement)
Daytona GraySort: 44.8 TB/min (old record: 15.9 TB/min; 2.8X greater performance)
Indy GraySort: 60.7 TB/min (old record: 18.2 TB/min; 3.3X greater performance)
Daytona MinuteSort: 37 TB/min (old record: 7.7 TB/min; 4.8X greater performance)
Indy MinuteSort: 55 TB/min (old record: 11 TB/min; 5X greater performance)