Tachyum Inc. announced that it has further expanded the capabilities of its Prodigy Universal Processor through support for TensorFlow and PyTorch environments, enabling a faster, less expensive and more dynamic solution for the most challenging artificial intelligence/machine learning workloads.
Analysts predict that AI revenue will surpass $300 billion by 2024 with a compound annual growth rate (CAGR) of up to 42 percent through 2027. AI is being heavily invested in by technology giants looking to make the technology more accessible for enterprise use-cases. They include self-driving vehicles to more sophisticated and control-intensive disciplines like Spiking Neural Nets, Explainable AI, Symbolic AI and Bio AI. When deployed into AI environments, Prodigy is able to simplify software processes, accelerate performance, save energy and better incorporate rich data sets to allow for faster innovation.
Recommended AI News: MedeAnalytics Features MedeCreate Healthcare Analytics Platform-as-a-Service at HLTH 2019
Proprietary programming environments like CUDA are inherently hard to learn and use. With open source solutions like TensorFlow and PyTorch, there are a hundred times more programmers that can leverage the frameworks to code for large-scale ML applications on Prodigy. By including support for deep learning environments that are easier to learn, build and train diversified neural networks, Tachyum is able to overcome and move beyond the limitations facing those working exclusively with NVIDIAs CUDA or with OpenCL.
In much the same way that external floating-point coprocessors and vector coprocessor chips have been internalized into the CPU, Tachyum is making external matrix coprocessors for AI an integral part of the CPU. By having integrated matrix operations as part of Prodigy, Tachyum is able to provide high-precision neural network acceleration of up to 10 times faster than other solutions. Tachyums support of 16-bit floating point and lower precision data types improves performance and saves energy in applications, such as video processing. Faster than the NVIDIA A100, Prodigy uses compressed data types to allow larger models to fit in memory. Instead of 20GB shared coherent memory, Tachyum allows 8TB per chip and 64TB per node.
Idle Prodigy-powered universal servers in hyperscale data centers, during off-peak hours, will deliver 10x more AI Neural Network training/inference resources than currently available, CAPEX free (i.e. at low cost, since the Prodigy-powered universal computing servers are already bought & paid for). Tachyums Prodigy enables edge computing and IOT products, which will have an onboard high-performance AI inference optimized to exploit Prodigy-based AI training from either the cloud or the home office.
Recommended AI News: BitPay Expands Integration with Coinbase to Enable Instant, No Fee Blockchain Payments
Business and trade publications are predicting just how important AI will become in the marketplace, with estimates of more than 50 percent of GDP growth coming from it, said Dr. Radoslav Danilak, Tachyum founder and CEO. What that means is that the less than 1 percent of data processed by AI today will grow to as much as 40 percent and the 3 percent of the planets power used by data centers will grow to 10 percent in 2025. There is an immediate need for a solution that offers low power, fast processing and easy of use and implementation. By incorporating open source frameworks like TensorFlow and PyTorch, we are able to accelerate AI and ML into the world with human-scale computing coming in 2 to 3 years.
Tachyums Prodigy can run HPC applications, convolution AI, explainable AI, general AI, bio AI and spiking neural networks, as well as normal data center workloads on a single homogeneous processor platform with its simple programming model. Using CPU, GPU, TPU and other accelerators in lieu of Prodigy for these different types of workloads is inefficient. A heterogeneous processing fabric, with unique hardware dedicated to each type of workload (e.g. data center, AI, HPC), results in underutilization of hardware resources, and a more challenging programming environment. Prodigys ability to seamlessly switch among these various workloads dramatically changes the competitive landscape and the economics of data centers.
Recommended AI News: Dataiku Raises $100 Million To Extend Leadership In Enterprise AI Market
Prodigy significantly improves computational performance, energy consumption, hardware (server) utilization and space requirements compared to existing chips provisioned in hyperscale data centers today. It will also allow Edge developers for IoT to exploit its low power and high performance, along with its simple programming model to deliver AI to the edge.
Prodigy is truly a universal processor. In addition to native Prodigy code, it also runs legacy x86, ARM and RISC-V binaries. And, with a single, highly efficient processor architecture, Prodigy delivers industry-leading performance across data center, AI, and HPC workloads. Prodigy, the companys flagship Universal Processor, will enter volume production in 2021. In April, the Prodigy chip successfully proved its viability with a complete chip layout exceeding speed targets. In August, the processor is able to correctly execute short programs, with results automatically verified against the software model, while exceeding the target clock speeds. The next step is to get a manufactured wholly functional FPGA prototype of the chip later this year, which is the last milestone before tape-out.
Prodigy outperforms the fastest Xeon processors at 10x lower power on data center workloads, as well as outperforming NVIDIAs fastest GPU on HPC, AI training and inference. A mere 125 HPC Prodigy racks can deliver 32 tensor EXAFLOPS. Prodigys 3X lower cost per MIPS and 10X lower core power translates to a 4X lower data center Total Cost of Ownership (TCO), enables billions of dollars of savings for hyperscalers such as Google, Facebook, Amazon, Alibaba, and others. Since Prodigy is the worlds only processor that can switch between data center, AI and HPC workloads, unused servers can be used as CAPEX-free AI or HPC cloud, because the servers have already been amortized.
Recommended AI News: Moonstake Web Wallet Provides Staking Support for Cardano ADA
Go here to see the original:
Tachyum Prodigy Native AI Supports TensorFlow and Pytorch - AiThority
- Calls to Ban Open Source are Misguided and Dangerous - The New Stack - June 26th, 2024
- Delving the Risks and Rewards of the Open-Source Ecosystem - InformationWeek - June 26th, 2024
- Enhancing security through collaboration with the open-source community - Help Net Security - June 18th, 2024
- It's time to face the open source security problem - ITPro - June 18th, 2024
- Mistral AI just launched 'Codestral', its own competitor to Code Llama and GitHub Copilot and it's fluent in over 80 ... - ITPro - June 2nd, 2024
- Open-source cybersecurity could derail the internet as we know it - Quartz - May 15th, 2024
- Developer Experience Influenced by Open Source Culture - InfoQ.com - May 15th, 2024
- BLint: Open-source tool to check the security properties of your executables - Help Net Security - May 15th, 2024
- Modular Open-Sources Mojo: The Programming Language that Turns Python into a Beast - MarkTechPost - April 2nd, 2024
- Meet the 21-Year-Old Creator of Devika, the Indian Open Source Devin Alternative - Analytics India Magazine - April 2nd, 2024
- Is Open Source Under Threat or Primed to Go to the Next Level? - The New Stack - March 13th, 2024
- Where is Technology Headed in 2024? - Open Source For You - March 13th, 2024
- A Detailed Conversation on Open-Source AI Frameworks for MLOps Workflows and Projects - AiThority - March 5th, 2024
- Everything you need to know about GitHub's new push protection changes - ITPro - March 5th, 2024
- StarCoder 2 is a code-generating AI that runs on most GPUs - TechCrunch - March 5th, 2024
- Is the future of open source software at risk due to protestware? - Tech Xplore - February 25th, 2024
- Google unveils new family of open-source AI models called Gemma to take on Meta and othersdeciding open-source AI aint so bad after all - Fortune - February 25th, 2024
- Jim Zemlin and the Linux Foundation share not-so-secret open-source sauce - ZDNet - February 25th, 2024
- Open source vs closed source AI: What's the difference and why does it matter? - Euronews - February 25th, 2024
- Biden administration to debate whether all AI systems should be open-source or closed - Firstpost - February 25th, 2024
- Some Linkerd service mesh users will soon have to pay - TechTarget - February 25th, 2024
- A lone developer just open sourced a tool that could bring an end to Nvidia's AI hegemony AMD financed it for ... - TechRadar - February 25th, 2024
- Scoping Out the Software-Defined Vehicle: The Benefits of OTA Updates & Open Source - Embedded Computing Design - February 25th, 2024
- The importance and limitations of open source AI models - TechTarget - February 9th, 2024
- 15+ Popular Python IDEs in 2024: Choosing The Best One - Simplilearn - February 9th, 2024
- Balancing Innovation and Security: The Open-Source Conundrum - BNN Breaking - February 9th, 2024
- VCs and startups love open-source AI models but how will they make money? - Sifted - February 9th, 2024
- How better and cheaper software could save millions of dollars while improving Canada's health-care system - The Conversation Indonesia - February 9th, 2024
- Best of 2023: Are We Witnessing the End of Open Source? - DevOps.com - December 28th, 2023
- What comes after open source? Bruce Perens is working on it - The Register - December 28th, 2023
- 200 GB of GTA 5 source code is about to get leaked, making it an open source: Report - Sportskeeda - December 28th, 2023
- Never was so much owed by so many to so few - a look at the unheralded heroes of the open source world - TechRadar - December 28th, 2023
- Rockstar hit with another cyberattack, leaked GTA 5 source code reveal cancelled DLC plans - Times of India - December 28th, 2023
- What is open source software? - Android Police - December 20th, 2023
- Feds Warn Health Sector to Watch for Open-Source Threats - BankInfoSecurity.com - December 11th, 2023
- OpenTofu: Open-source alternative to Terraform - Help Net Security - December 11th, 2023
- AWS exec: 'Our understanding of open source has started to change' - The Register - December 11th, 2023
- Mark Jelic Rings in 40 Years Since the TEC-1 Launch with a New, Open Source, Upgraded TEC-1G SBC - Hackster.io - December 11th, 2023
- AI's future could be 'open-source' or closed. Tech giants are divided as they lobby regulators - Tech Xplore - December 11th, 2023
- Cyber Security Today, Nov. 24, 2023 A warning to tighten security on Kubernetes containers, and more - IT World Canada - November 25th, 2023
- This AI Paper Proposes ML-BENCH: A Novel Artificial Intelligence Approach Developed to Assess the Effectiveness of LLMs in Leveraging Existing... - November 25th, 2023
- Generative AI is a genuine breakthrough unlike most fads in tech: Zerodha CTO Kailash Nadh on the current waves in tech - The Hindu - October 27th, 2023
- Meet RedPajama: An AI Project to Create Fully Open-Source Large Language Models Beginning with the Release of a 1.2 Trillion Token Dataset -... - April 25th, 2023
- Hashtag Trending Apr.24th- Cybersecurity workers burnout; Code generated by ChatGPT and Googles Bard not very secure; Execs would want a robot to make... - April 25th, 2023
- This AI Project Brings Doodles to Life with Animation and Releases Annotated Dataset of Amateur Drawings - MarkTechPost - April 17th, 2023
- EU shares best practices with Ukrainian law enforcers on Open Source Intelligence and Criminal Analysis to - EIN News - April 8th, 2023
- 'I've never seen anything like this:' One of China's most popular apps has the ability to spy on its users, say experts - CNN - April 8th, 2023
- With Just ~20 Lines of Python Code, You can Do Retrieval Augmented GPT Based QA Using This Open Source Repository Called PrimeQA - MarkTechPost - March 5th, 2023
- Daily Crunch: Hundreds of Salesforce workers laid off in January just discovered they were out of work today - TechCrunch - February 7th, 2023
- Unlocking the power of Open AI: how to automate information extraction - The Hindu - February 7th, 2023
- Is composable business most essential technology trend to meet challenges of 2023 and beyond? - ComputerWeekly.com - January 30th, 2023
- Open Definition & Meaning | Dictionary.com - January 22nd, 2023
- 529 Synonyms & Antonyms of OPEN - Merriam-Webster - January 22nd, 2023
- Open Definition & Meaning - Merriam-Webster - January 22nd, 2023
- Can Wazuh Become The Worlds Largest Open Source Cybersecurity Platform And IPO Without VC Funding? - Forbes - January 6th, 2023
- 8 Free/Open Source Code Review Tools for 2022 - SoftwareSuggest - December 28th, 2022
- Finding the next Log4j OpenSSFs Brian Behlendorf on pivoting to a risk-centred view of open source development - The Daily Swig - December 28th, 2022
- Nithin Kamath says FOSS is the 'pillar' on which Zerodha has been built. What is it? - Business Today - December 28th, 2022
- How Dogeliens Will Take Over the Metaverse Like Bitcoin and Stellar Took Over the Crypto World. - newsbtc.com - December 28th, 2022
- Intrinsic Buys Open Robotics' Commercial Arm, But Leaves ROS and Gazebo with the Foundation - Hackster.io - December 20th, 2022
- Open-source code is everywhere; GitHub expands security tools to help ... - December 20th, 2022
- Security Of Enterprise Code: What Companies Using Open-Source Software Should Know About Binary Code Verification - Forbes - December 20th, 2022
- Open Source - Apple Developer - December 12th, 2022
- Your Code of Conduct | Open Source Guides - December 12th, 2022
- Code of Conduct | Meta Open Source - Facebook - December 12th, 2022
- From the creator of Homebrew, Tea raises $8.9M to build a protocol that helps open source developers get paid - TechCrunch - December 12th, 2022
- Consortium of Japan partners successfully promote domestic production and cost reduction for 5G core technology, the basis for next-generation... - November 25th, 2022
- GitHub Vulnerability Allows Hackers to Hijack Thousands of Popular Open-Source Packages - CPO Magazine - November 17th, 2022
- GitHubs Octoverse report finds 97% of apps use open source software - VentureBeat - November 17th, 2022
- Microsoft sued for open-source piracy through GitHub Copilot - BleepingComputer - November 7th, 2022
- The White House Memorandum on Securing the Software Supply Chain: What It Means for Your Organization - Security Boulevard - November 7th, 2022
- First Timers Only - Get involved in Open Source and commit code to your ... - October 23rd, 2022
- List of free and open-source software packages - Wikipedia - October 23rd, 2022
- What is open source? - Red Hat - October 23rd, 2022
- Introducing Triton: Open-Source GPU Programming for Neural Networks - October 23rd, 2022
- Comparison of open-source and closed-source software - October 23rd, 2022
- Java 19 Brings New Patterns to Open Source Programming Language - October 23rd, 2022
- API series - OctoML: ML APIs need to take a lesson from their ancestors - ComputerWeekly.com - October 23rd, 2022
- Benefits of working with open source data quality solutions - TechRepublic - October 15th, 2022
- Microsoft's GitHub Copilot AI is making rapid progress. Here's how its human leader thinks about it - CNBC - October 15th, 2022