What Will Watson Do with a Power8 Brain?

IBM’s Watson impressed people all over the world in 2011 when the machine beat all contenders in a game of Jeopardy! Since then, IBM has kept the Power7-based technology busy, with gigs in financial analysis, healthcare, and customer service, among…

IBM’s Watson impressed people all over the world in 2011 when the machine beat all contenders in a game of Jeopardy! Since then, IBM has kept the Power7-based technology busy, with gigs in financial analysis, healthcare, and customer service, among others. But now that IBM is gearing up to ship its Power8 processors, we could see a newer and more powerful Watson emerge.

IBM unveiled details of the forthcoming Power8 processor this week at the Hot Chips conference at Stanford University. The 12-core, 4Ghz Power8 chip can execute 96 threads simultaneously, and is expected to be 2x to 3x more powerful than IBM’s Power7 chip, which was used in the Watson supercomputer that competed on the game show.

The Power8’s capability to move data has been dramatically improved over previous designs, including the interim Power7+ chip that IBM shipped in 2012. According to the Hot Chips presentation on Monday by Jeff Stuecheli, the chief nest architect for Power8, the new processor features 230 GB per second of sustained memory bandwidth between the L4 caches and the processors, and total peak I/O of 48 GB per second along the new on-die integrated PCI-Express 3.0 controllers, making it a very speedy chip.
One of the cool things IBM has done with those integrated PCI-Express 3.0 controllers is to design a new transport layer called the Coherence Attach Processor Interface (CAPI) that will enable co-processors, including GPUs or field programmable gate arrays (FPGAs) to be connected directly to the Power8 chip and to share data stored in its memory. Nvidia, which together with Google and Mellanox signed onto the OpenPOWER Consortium that IBM unveiled earlier this month, is expected to build a GPU that plugs into the Power8 chips through this CAPI protocol.
So, what does all this mean for Watson, the 2,880-core supercomputer that stole so many hearts during Jeopardy!? On a purely thread-count and parallel workload basis, the Power8 processor should be a boon for Watson, which, with Power7 technology, was able to process 500 gigabytes of information—the equivalent of a million books—every second.
The truly intriguing part, however, is what a new Power8-based supercomputer combined with Nvidia GPUs and a Mellanox interconnect can do for HPC workloads, including the DeepQA (question answering) technology that Watson is based on.
According to IBM, DeepQA combined uses more than 100 different techniques in its various workload components, including natural language processing, machine learning, hypothesis generation, evidence gathering, analysis, and scoring. Apache’s Hadoop and UIMA (Unstructured Information Management Architecture) frameworks also played a role in Watson’s software.
The area where Watson struggled the most—correctly assessing contextual clues to determine the true meaning of a statement—is a particularly difficult problem that artificial intelligence and machine learning experts have yet to crack. Perhaps the Power8 technology, combined with developments from the OpenPOWER Consortium, will help find better answers.
“Watson wasn’t a traditional workload for us,” IBM’s Stuecheli said during his Hot Chips presentation, according to a PC World story. “We’d like to find more of these opportunities.”

Sponsored Links

Realize the potential of Hadoop™ with the Cray CS300™ Cluster supercomputerRun complex algorithms in minutes not hours with Cray optimized turnkey Hadoop solution for Big Data Analytics

RSS Feeds

Feeds by Topic
Feeds by Industry

Feeds by Content Type
Subscribe to All Content


Short TakesMoore’s Law We Miss You AlreadyAug 29, 2013 |
As transistors reach the limits of miniaturization, it is only a matter of time until Moore’s Law runs out of steam. The latest expert to weigh-in says Moore’s Law will expire in 2020 at the 7nm node.Read more…

ADEPT Emphasizes Energy-Efficient ParallelismAug 29, 2013 |
The EU-funded ADEPT project is exploring the energy-efficient use of parallel technologies by combining the talents of HPC and the embedded sector. The goal is to develop a tool for modeling and predicting the power and performance characteristics of parallel systems.Read more…

Biotech Vendors Compute to CompeteAug 28, 2013 |
Biotech companies are reaping the benefits of more accessible high-performance computing tools. The ability to compress weeks of modeling into days or hours is improving outcomes and boosting time to market.Read more…

The Evolving Memory MarketAug 28, 2013 |
The market for computer memory is entering a period of punctuated evolution as a result of several forces, including the continued growth of mobile devices like smartphones and tablets, as well as growth in the cloud data centers and communication networks that serve data to mobile users. HPC workloads also play a part in the changing memory landscape.Read more…

EGI Cloud: SLAs and Summer SchoolAug 28, 2013 |
The EGI Federated Cloud is a new kind of research e-infrastructure designed to serve the scientific community. In support of this mission, the EGI has contributed to a report on service level agreements and also provided hands-on training.Read more…

Read more headlines…

Sponsored Whitepapers

Extracting Maximum ROI from HPC Server Technologies

08/29/2013 | Asetek, Cray, HP, Sabalcore and Supermicro | Learn a number of methods by which HPC users can increase the ROI of their installations which include:Software technologies for ROI gain
Server hardware technologies enabling greater efficiency
Data center power and cooling methodology for cost savings

Additional discussion is around ROI improvement coming from cloud computing solutions as a component of an HPC workload.Technical Computing for a New Era

07/30/2013 | IBM | This white paper examines various means of adapting technical computing tools to accelerate product and services innovation across a range of commercial industries such as manufacturing, financial services, energy, healthcare, entertainment and retail. No longer is technically advanced computing limited to the confines of big government labs and academic centers. Today it is available to a wide range of organizations seeking a competitive edge.

View the White Paper Library

Sponsored Multimedia

Xyratex, presents ClusterStor at the Vendor Showdown at ISC13

Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
HPCwire Live! Atlanta’s Big Data Kick Off Week Meets HPC

Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta’s first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today’s big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?

More Multimedia

HPC Job Bank

Visit the HPCwire Job Bank

Featured Events

August 7, 2013
– September 04, 2013

The 2013 HPCwire Readers’ Choice Awards Nominations are now open! San Jose ,

September 9, 2013
– September 11, 2013

HPC User Forum US MeetingBoston, MAUnited States

September 9, 2013
– September 09, 2013

10th Annual HPC for Wall StreetNew York City, NYUnited States

September 12, 2013
– September 12, 2013

HPC Advisory Council Spain Conference 2013Barcelona, Spain

September 25, 2013
– September 25, 2013

Cyber Security SummitNYC, NYUnited States

November 17, 2013
– November 22, 2013

SC’13Denver, COUnited States
View/Search Events
Post an Event