Business challenge

The engineers of the IBM Systems High Speed Bus Signal Integrity (HSB-SI) Team are challenged to run design validation simulation analysis of high-speed interfaces for the purpose of choosing an optimal configuration point, in minimum time.

Transformation

To design the future’s world class leading server computer interfaces, the HSB-SI Team implemented the IBM Bayesian Optimization (IBO) software, a machine learning tool developed by IBM Research, reducing the number of simulations needed to reach the optimal configuration point for chip-to-chip communication.

Results

140x

faster time to solution with higher accuracy than legacy method

99%

less cores utilized to arrive at a higher confidence solution

Less than 1%

error in solution using IBO

Business challenge story

Chip-to-chip communication

Did you know that the IBM High Speed Bus Signal Integrity (HSB-SI) Team significantly contributed to the design and development of the world’s two fastest supercomputers, Summit (#1) and Sierra (#2)?

The HSB-SI team is comprised of engineers that do high speed bus signal integrity design and analysis for server systems using the IBM POWER Processors. A key factor giving these supercomputers a performance advantage is being able to run the communication between chips at the fastest speed possible. The HSB-SI team designs the route that the signal takes from one chip to another, so that the signal sent from the transmitter and arriving at the receiver can be deciphered with no errors. In order to design or analyze the signal integrity of a channel between two chips, these engineers may need to do numerous simulations, that can take several days of work towards ensuring error free communication at maximum speed.

As Jose Hejase, Ph.D., Senior Engineer, High Speed Bus Signal Integrity Analysis at IBM, explains “To guarantee error free communication, the SI engineer has to make difficult decisions about design constraints on a channel. How to design the printed circuit board? Should I use connectors? What type of cables should I use? Will the channel still work under worst case manufacturing tolerances? Each design is unique and there is no single recipe that answers all the questions. Doing this analysis using our traditional means can be very time consuming, tedious, and engineering intensive.”

Every day, companies that design electronics call upon their Signal Integrity (SI) teams to make difficult decisions about design options and this SI team needed a new solution to free up their engineers from days of simulations to obtain optimal channels for very low bit error rate (practically error free) chip-to-chip communication.

Transformation Story

Error free communication

 

 

 

 

 

IBM Bayesian Optimization vs. Traditional

The engineers on the HSB-SI team are challenged to ensure communication between chips at maximum speed. The traditional process to analyze these design channels are engineering intensive and can take up to several days of work before arriving at an optimal channel design combination. The HSB-SI team realized they had an opportunity to implement a design acceleration software tool developed by IBM Research known as the IBM Bayesian Optimization (IBO) software.

IBO is the optimal tool for the HSB-SI team. “Using IBO is a way to greatly reduce the number of simulations needed to come to an optimal point and free up the engineers to think of how to design the systems of the future now that they will no longer need to do less tedious work using traditional methods.” – Jose Hejase, Ph.D.

IBO is a statistical framework that uses machine learning to model and minimize an arbitrary objective function. As Jose Hejase, Ph.D., explained “In order to go through all of the channel design combinations and/or tolerances. The engineer has to do substantial thinking and planning to determine a case study to analyze. This analysis consists of multiple simulations and these simulations can take many days. This is where IBO comes in. IBO is an optimization tool that is driven by machine learning. Instead of us having to do an intensive sweep of all the different combinations and/or tolerances to be considered and the simulation tool settings. IBO is able to learn from a relatively small amount of simulation iterations to arrive at the optimal result for a particular channel.”

IBO has given us the opportunity to make our team more efficient in the highly computer intensive analysis we need to produce our systems.

Dale Becker, Ph.D, Chief Engineer, Electrical Packaging Integration, IBM, IBM

Results story

Discovering optimal designs faster

As for next steps, the HSB-SI team is working to incorporate IBO into their mainstream process as they continue to produce world changing computer systems.

In addition, IBO will allow engineers to do extensive tolerance analysis for high speed bus communication channels. This is important in order to ensure that channel designs will still perform well even under worst case manufacturing tolerances. “When designing a channel, there are multiple channel design options to consider. Within those channel designs are tolerances. Each individual segment of a particular channel will have tolerance associated with it. Once you sort all the tolerances together you might have more than 1,000 channel combinations. This computation is ideal for IBO. Instead of the engineer having to intensively go through each combination and determine the performance metric, IBO will do the optimization to arrive at the channel worst case performance under tolerance with high confidence.” – Jose Hejase, Ph.D.

IBO is a flexible and fast tool that has saved the compute time for the HSB-SI team and still allowed them to get high confidence channel designs. Moreover, IBO’s generated, machine learned models are potential vehicles to decrease dependence on certain simulation tools in specific scenarios to analyze channel performance. Last but not least, one benefit from IBO is that the sensitivity of channel performance to certain channel component design properties can be learned thus providing engineers hints on where to spend most effort in future designs for best SI performance.

“Machine learning and optimization is a hot area in the electronic design industry. In some way or another companies are moving towards that direction. Having IBO within IBM is a great advantage for us at IBM.” – Jose Hejase, Ph.D.

Our team is taking advantage of state-of-the-art machine learning to design computer systems of the future.

Dale Becker, Ph.D., Chief Engineer, Electrical Packaging Integration , IBM

About The IBM Power Systems Hardware Development Team

IBM Power Systems is a family of servers fueled by POWER processor technology and software built to crush clients’ most advanced data application-from the mission-critical workloads they run today to the next generation of AI.

The Power Systems hardware development team is a part of the team that design IBM systems including Power Systems, z Systems, and IBM Storage. This team designs the power, packaging, and cooling for Power Systems including the current top two supercomputers globally, Summit and Sierra.

Take the Next Step

To learn more about IBM Power Systems, please contact your IBM representative or IBM Business Partner, or visit the following website:  https://www.ibm.com/it-infrastructure/power