The two chip companies have signed a term sheet, according to sources with direct knowledge of the agreement. Intel has signed a term sheet to acquire the AI chip startup SambaNova Systems, two sources with direct knowledge of the agreement tell WIRED. The details of the term sheet are unknown. The agreement is non-binding, meaning the deal is not yet finalized and could be dissolved without penalty. It could take weeks or even months before regulatory approval, liability scrutiny, and financial due diligence are complete.

Mapping a dataflow - graph of an ML model onto a reconfigurable system is difficult, as different mappings have different throughputs and consume resource constraints differently. To solve this, a model to evaluate the throughput of mappings is necessary as measuring throughput completely is expensive. Many use a hand - designed analytical model, relying on proxy features or intuition, introducing error. We provide a Learned Approach that predicts throughput 31% - 52% more accurately over a variety of graphs. In addition, our approach shows no accuracy degradation after removing performance annotations. We show that using this approach results in 5.6% faster compiled graphs.

Token generation speed is critical to power the next wave of AI inference applications. GPUs significantly underperform during token generation due to synchronization overheads at kernel boundaries, utilizing only 21% of their peak memory bandwidth. While recent dataflow architectures mitigate these overheads by enabling aggressive fusion of decoder layers into a single kernel, they too leave performance on the table due to synchronization penalties at layer boundaries. This paper presents kernel looping, a specialized global optimization technique which exploits an optimization opportunity brought by combining the unique layer-level fusion possible in modern dataflow architectures with the repeated layer structure found in language models. Kernel looping eliminates synchronization costs between consecutive calls to the same kernel by transforming these calls into a single call to a modified kernel containing a pipelined outer loop. We evaluate kernel looping on the SambaNova SN40L Reconfigurable Dataflow Unit (RDU), a commercial dataflow accelerator for AI. Experiments demonstrate that kernel looping speeds up the decode phase of a wide array of powerful open-source models by up to 2.2$\times$ on SN40L. Kernel looping allows scaling of decode performance over multiple SN40L sockets, achieving speedups of up to 2.5$\times$. Finally, kernel looping enables SN40L to achieve over 90% of peak performance on 8 and 16 sockets and achieve a speedup of up to 3.7$\times$ over DGX H100. Kernel looping, as well as the models evaluated in this paper, are deployed in production in a commercial AI inference cloud.

Most forms of technology applications are well understood. Every computer programme can be deconstructed into the basic building blocks of code, and if it goes wrong, you can debug the software – often by simply stepping through the code line by line in order to find out where the problem lies. Artificial Intelligence, or AI, is different. With the latest AI large language models we can't predict exactly what it will output, but it will do a good job at writing an article or creating poetry. What makes them human-like is the lack of predictable outcomes – humans simply aren't predictable!

OTP Bank has selected SambaNova Systems to help build Europe's fastest AI Supercomputer, positioning OTP Group and the Ministry for Innovation and Technology (MIT) of Hungary as European leaders in the modern day space race to a post AI business. The partnership and plan is to create the fastest AI supercomputer in Europe in the coming 100 days in an attempt to leapfrog both Wall Street and Western European competitors in the AI race. OTP Bank is looking to produce an AI system to be used for national research. This research will help to aid the private and public sectors, as well as higher education across all of Central and Eastern Europe in cooperation with the ITM of Hungary. When speaking to Co-founder and CEO of SambaNova Systems, Rodrigo Liang, alongside Marshall Choy, Senior Vice President and Head of Product at SambaNova Systems, it became easier to understand the scale of the project at hand.

All the sessions from Transform 2021 are available on-demand now. SambaNova Systems, winner of VentureBeat's 2021 Innovation in Edge award, is a significant contender in the global edge computing market. The startup raised $676 million in April 2021 and is moving from its origins as an AI-specific chip company to one that provides comprehensive dataflow-as-a-service to its clients. Research firm MarketsandMarkets forecasts an impressive 34% compounded annual growth rate for the market and anticipates its value reaching $15.7 billion by 2025. Traditional central processing units (CPUs) and graphics processing units (GPUs) are based on transactional processing, which needs accuracies to the nth degree for computations.

ScaleWorX, which provides businesses with infrastructure solutions for data-intensive computing and AI, will now offer innovative and best-in-class AI …

Lawrence Livermore National Laboratory (LLNL) has installed a state-of-the-art artificial intelligence (AI) accelerator from SambaNova Systems, the National Nuclear Security Administration (NNSA) announced today, allowing researchers to more effectively combine AI and machine learning (ML) with complex scientific workloads. LLNL has begun integrating the new AI hardware, SambaNova Systems DataScale, into the NNSA's Corona supercomputing cluster, an 11-plus petaFLOP machine that Lab scientists are using to conduct fusion energy research for stockpile stewardship applications, find therapeutics for COVID-19 and perform other unclassified basic science work. Lab researchers said the upgrade will allow them to run scientific simulations on the Corona system while offloading AI calculations from those simulations to the SambaNova DataScale system, improving overall speed, performance and productivity. "This integration enables low-latency communication between the two devices allowing them to operate in tandem with greater overall efficiency," said LLNL computer scientist Ian Karlin, who heads the SambaNova project. "In addition, scientific simulations running on Corona will feed data as they run into the SambaNova DataScale system to train new machine learning models based on their results."

A lot of the people who are working at the many AI chip startups have a long history in processor development in the datacenter, and that is certainly true of the folks who founded SambaNova Systems. And this is a fortunate thing because these people can leverage some of the good ideas they know worked when commercializing a new technology and avoid some of the big mistakes their former employers sometimes made. At our recent The Next AI Platform event, we sat down with Rodrigo Liang, co-founder and chief executive officer of SambaNova Systems, which is one of the upstart custom AI chip producers vying for attention and budget dollars. SambaNova was founded in 2017 by a bunch of ex-Sun Microsystems techies as well as a few from Stanford University, which is of course where Sun itself was born in 1982. The co-founders include Kunle Olukotun and Chris Ré, professors at Stanford, with Olukotun being the leader of the Hydra chip multiprocessor research project and sometimes known as the father of the multicore processor.

Much of what sets The Next Platform apart from other tech publications is depth and analysis. As it turns out, the key to getting both of those facets is knowing what questions to ask and pushing for answers that go beyond the basic and cut through marketing and hype. This time we are conducting interviews in a new format--and we want you involved in the process. Please join us on March 10, 2020 at The Glasshouse in downtown San Jose, CA for an all-day event featuring the same in-depth conversations you expect from TNP (and from our sold-out Next AI Platform event last year), live on-stage followed by a cocktail reception and evening dinner opportunities for networking with key people defining the next generation of AI infrastructure. Meet the Next Platform team with plenty of time to talk about what matters to you, get first access to exclusive interviews, and spend the day with us in an intimate setting at San Jose's premier event venue, The Glasshouse. Just some of the best interviewers in the high-end infrastructure space and a lineup of thought leaders building the next generation of large-scale infrastructure to support emerging AI workloads.