Start-up Aims AI at Automated Tuning of Complex Systems

By Doug Black

February 22, 2018

Today’s bigger, more complex, connected and intelligent systems have an exponentially higher number of connections, dependencies, interfaces, protocols and processing architectures that, if not optimized, will hamstring performance. For performance engineers still using manual tuning methods, systems optimization is a time-eating hydra-headed monster that poses a virtually infinite number of possible adjustments and configurations for trial-and-error testing.

Enter AI.

Start-up Concertio today launched what it said is the first machine learning-based tool aimed at making applications and systems operate for maximized performance by optimizing the myriad of configuration settings used in complex systems. While its initial product, Optimizer Studio, automates systems diagnostics and generates a “grocery list” of adjustments that systems engineers and IT managers then review and implement, the next iteration of the technology (Optimizer Runtime) will automate both diagnostics and tuning implementation.

None less than Mellanox has tested Concertio’s effectiveness on its own networking technology. As part of today’s launch, Concerto announced the results of a test involving Mellanox’s ConnectX-3 Pro Ethernet cards that compared performance after automated testing by Optimizer Studio against manual tuning methods used by Mellanox performance engineers.

The network cards were delivered in their off-the-shelf default settings, and then the Mellanox engineers and Concertio were informed of the workload the cards would be used for. Optimizer Studio ran against nine ConnectX-3 Pro specific knobs representing millions of option combinations. The tool’s workload classification engine and reinforcement learning techniques modeled the target workload, detecting different workload phases, and experimenting with various knob configurations in each phase. It then produced a report showing the optimal settings for the specific use-case.

The test result: Concertio won.

“In the comparison test, Optimizer Studio’s automated run improved performance in the target use-case by 80 percent, surpassing the 62 percent we achieved by manual tuning,” said Amir Ancel, performance group director at Mellanox. “Optimizer Studio’s automated tuning algorithms allow us to focus on high-level optimization, leaving the mundane low-level parameter optimizations to software.”

Built for traditional datacenters, hyperscale datacenters and high-performance computing systems in the cloud or on-premises, Optimizer Studio monitors and learns from the interactions between applications and systems, according to Concertio.

As of now, the tools support Linux-based systems running on Intel CPUs. Concertio said it intends to broaden its portfolio of supported technologies in upcoming product iterations.

“Tuning used to be easy,” Dr. Tomer Morad, Concertio co-founder and CEO, shared. “There were only a handful of knobs, and you’d put a performance engineer on that and tweak some settings and get some good results. But today we are already in the hundreds of knobs. It’s exploding and become almost impossible to get to a very good result because the parameter space is practically limitless. If you have 100 binary knobs it’s practically limitless, you cannot check everything. So you need some kind of automatic tool to help you with that.”

Tuning variables, called tunable knobs, include settings across hardware, firmware, the operating system and applications such as:

  • CPU hardware, including symmetric multi-threading, cache prefetching and cache partitioning configuration; peripheral hardware, including PCIe maximum read request size, network interrupt affinity and network task offloading
  • Firmware, including power states of the CPU
  • Operating system, including choice of IO or task scheduler, NUMA balancing and memory migration, thread affinity and page cache
  • Applications, including application framework settings (e.g., Spark), application component (e.g., MongoDB database) settings, and application-defined knobs

With so many variables, Morad said, performance engineers can’t be expected to know about all of the available knobs, and to predict their effects on one another. IT professionals must also occasionally tune their systems, but it’s difficult for them to maintain experts on all system internals, he said. Too often, engineers test and set optimal settings for a few knobs they are more familiar with, leaving the rest in default settings that were in place when the equipment was delivered. In some cases, he said, system tuning is overlooked entirely due to the complexity involved, and systems remain at inefficient under-performing factory settings.

Beyond application performance tuning, Morad said the tool can be used for cutting cloud and data center costs by finding system configurations that use fewer resources, and it can be utilized by hardware and software product vendors to identify optimized off-the-shelf configurations for shipment to customers or reseller. It also can be used for maximizing public benchmark performance for marketing purposes (including LINPACK).

Privately-held Concertio (previously called DatArcs) was founded in 2016 and is based in New York City. The company is part of the Runway Program at the Jacobs Technion-Cornell Institute of Cornell Tech in New York, and it’s with the Intel Ingenuity Partner Program.

“Tailor-tuned systems can significantly outperform baseline general-purpose systems, but the number of configurable settings has reached into the hundreds – way too many for any human team to effectively tune and test,” said Concertio co-founder and CTO Andrey Gelman. “It used to be merely a gap that could be bridged by human tuning and testing, but with the increasing hardware and software complexity, it’s exploding into a chasm where human performance tuning is netting diminishing returns, leaving these expensive systems bottlenecked and inefficient.”

A version of this article originally appeared on our sister site, EnterpriseTech.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire