Nervana Systems

Last updated
Nervana
Company type Subsidiary of Intel
Industry Artificial intelligence
Founded2014
Headquarters San Diego, California, U.S. and Palo Alto, California, U.S.
Number of employees
48 (2016)  OOjs UI icon edit-ltr-progressive.svg
Website nervanasys.com   OOjs UI icon edit-ltr-progressive.svg

Nervana Systems was an artificial intelligence software company based in San Diego, California, and Palo Alto, California. [1] The company provided a full-stack software-as-a-service platform called Nervana Cloud that enabled businesses to develop custom deep learning software. [2] On August 9, 2016, it was acquired by Intel, for an estimated $408 million. [3] [4]

Contents

Deep learning framework

The company's (now discontinued) open source deep learning framework is called neon. [5] Neon – which the company said would outperform rival frameworks such as Caffe, Theano, Torch, and TensorFlow [5] – would achieve its performance advantage through assembler-level optimization, multi-GPU support, and use of an algorithm called Winograd for computing convolutions, which are common mathematical operations in the deep learning process. [6]

Nervana Cloud

Nervana Cloud, [7] announced in February 2016, is based on Neon, and ran on Nvidia Titan X GPUs, but Nervana was also developing a custom application-specific integrated circuit (ASIC) called the Nervana Engine that was optimized for deep learning and that Nervana said would perform 10x better than Nvidia Maxwell architecture GPUs. The Nervana Engine was expected to achieve greater compute density by implementing only those design elements that are necessary to support deep learning algorithms and ignoring legacy elements specific to graphics processing. [8]

History

Nervana was founded in 2014 by CEO Naveen Rao, CTO Amir Khosrowshahi (Dara Khosrowshahi's cousin [9] ), and VP Algorithms Arjun Bansal. [10] Nervana has raised $28 million in funding. [1] In June 2015, Nervana raised $20.5 million in series A funding led by Data Collective with participation from Allen & Company, AME Cloud Ventures, Playground Global, the CME Group, Draper Fisher Jurvetson, Fuel Capital, Lux Capital, and Omidyar Network. [11] It was estimated to have only 48 employees when it was acquired by Intel in 2016. [3] In January 2020, Intel shut down the development of Nervana in favor of its acquisition of Habana Labs. [12]

Related Research Articles

<span class="mw-page-title-main">Nvidia</span> American multinational technology company

Nvidia Corporation is an American multinational corporation and technology company headquartered in Santa Clara, California, and incorporated in Delaware. It is a software and fabless company which designs and supplies graphics processing units (GPUs), application programming interfaces (APIs) for data science and high-performance computing as well as system on a chip units (SoCs) for the mobile computing and automotive market. Nvidia is also a dominant supplier of artificial intelligence (AI) hardware and software.

<span class="mw-page-title-main">Graphics processing unit</span> Specialized electronic circuit; graphics accelerator

A graphics processing unit (GPU) is a specialized electronic circuit initially designed to accelerate computer graphics and image processing. After their initial design, GPUs were found to be useful for non-graphic calculations involving embarrassingly parallel problems due to their parallel structure. Other non-graphical uses include the training of neural networks and cryptocurrency mining.

Tilera Corporation was a fabless semiconductor company focusing on manycore embedded processor design. The company shipped multiple processors in the TILE64, TILEPro64, and TILE-Gx lines.

<span class="mw-page-title-main">Arm Holdings</span> British multinational semiconductor and software design company

Arm Holdings plc is a British semiconductor and software design company based in Cambridge, England, whose primary business is the design of central processing unit (CPU) cores that implement the ARM architecture family of instruction sets. It also designs other chips, provides software development tools under the DS-5, RealView and Keil brands, and provides systems and platforms, system-on-a-chip (SoC) infrastructure and software. As a "holding" company, it also holds shares of other companies. Since 2016, it has been majority owned by Japanese conglomerate SoftBank Group.

Cloud gaming, sometimes called gaming on demand or game streaming, is a type of online gaming that runs video games on remote servers and streams the game's output directly to a user's device, or more colloquially, playing a game remotely from a cloud. It contrasts with traditional means of gaming, wherein a game is run locally on a user's video game console, personal computer, or mobile device.

<span class="mw-page-title-main">ServiceNow</span> American technology company

ServiceNow, Inc. is an American software company based in Santa Clara, California, that develops a cloud computing platform to help companies manage digital workflows for enterprise operations. Founded in 2003 by Fred Luddy, ServiceNow is listed on the New York Stock Exchange and is a constituent of the Russell 1000 Index and S&P 500 Index. In 2018, Forbes magazine named it number one on its list of the world's most innovative companies.

Intel Capital is a division of Intel Corporation, set up to manage corporate venture capital, global investment, mergers and acquisitions. Intel Capital makes equity investments in a range of technology startups and companies offering hardware, software, and services targeting artificial intelligence, autonomous technology, data center and cloud, 5G, next-generation compute, semiconductor manufacturing and other technologies.

Eclipse Deeplearning4j is a programming library written in Java for the Java virtual machine (JVM). It is a framework with wide support for deep learning algorithms. Deeplearning4j includes implementations of the restricted Boltzmann machine, deep belief net, deep autoencoder, stacked denoising autoencoder and recursive neural tensor network, word2vec, doc2vec, and GloVe. These algorithms all include distributed parallel versions that integrate with Apache Hadoop and Spark.

Kinetica is a distributed, memory-first OLAP database developed by Kinetica DB, Inc. Kinetica is designed to use GPUs and modern vector processors to improve performance on complex queries across large volumes of real-time data. Kinetica is well suited for analytics on streaming geospatial and temporal data.

Playground Global is an early-stage venture capital firm that invests in deep tech and assists startups with software, hardware, machine learning, marketing, talent and design. The company was founded in 2015 by Andy Rubin, Peter Barrett, Matt Hershenson and Bruce Leak. Playground offers startups support with engineering, distribution, manufacturing and financing in exchange for equity.

An AI accelerator, deep learning processor, or neural processing unit (NPU) is a class of specialized hardware accelerator or computer system designed to accelerate artificial intelligence and machine learning applications, including artificial neural networks and machine vision. Typical applications include algorithms for robotics, Internet of Things, and other data-intensive or sensor-driven tasks. They are often manycore designs and generally focus on low-precision arithmetic, novel dataflow architectures or in-memory computing capability. As of 2024, a typical AI integrated circuit chip contains tens of billions of MOSFET transistors.

Chainer is an open source deep learning framework written purely in Python on top of NumPy and CuPy Python libraries. The development is led by Japanese venture company Preferred Networks in partnership with IBM, Intel, Microsoft, and Nvidia.

<span class="mw-page-title-main">Graphcore</span> British semiconductor company

Graphcore Limited is a British semiconductor company that develops accelerators for AI and machine learning. It has introduced a massively parallel Intelligence Processing Unit (IPU) that holds the complete machine learning model inside the processor.

The bfloat16 floating-point format is a computer number format occupying 16 bits in computer memory; it represents a wide dynamic range of numeric values by using a floating radix point. This format is a shortened (16-bit) version of the 32-bit IEEE 754 single-precision floating-point format (binary32) with the intent of accelerating machine learning and near-sensor computing. It preserves the approximate dynamic range of 32-bit floating-point numbers by retaining 8 exponent bits, but supports only an 8-bit precision rather than the 24-bit significand of the binary32 format. More so than single-precision 32-bit floating-point numbers, bfloat16 numbers are unsuitable for integer calculations, but this is not their intended use. Bfloat16 is used to reduce the storage requirements and increase the calculation speed of machine learning algorithms.

DeepScale, Inc. was an American technology company headquartered in Mountain View, California, that developed perceptual system technologies for automated vehicles. On October 1, 2019, the company was acquired by Tesla, Inc.

HEAVY.AI is an American-based software company, that uses graphics processing units (GPUs) and central processing units (CPUs) to query and visualize big data. The company was founded in 2013 by Todd Mostak and Thomas Graham and is headquartered in San Francisco, California.

oneAPI (compute acceleration) Open standard for parallel computing

oneAPI is an open standard, adopted by Intel, for a unified application programming interface (API) intended to be used across different computing accelerator (coprocessor) architectures, including GPUs, AI accelerators and field-programmable gate arrays. It is intended to eliminate the need for developers to maintain separate code bases, multiple programming languages, tools, and workflows for each architecture.

DeepMap, is a Palo Alto, California-based software company that develops high definition (HD) maps for self-driving vehicles.

Christofari — are Christofari (2019), Christofari Neo (2021) supercomputers of Sberbank based on Nvidia corporation hardware Sberbank of Russia and Nvidia. Their main purpose is neural network learning. They are also used for scientific research and commercial calculations.

References

  1. 1 2 "Nervana Systems Puts Deep Learning AI in the Cloud". IEEE Spectrum: Technology, Engineering, and Science News. Retrieved 2016-06-22.
  2. ↑ "Nervana Systems Brings Deep Learning to the Masses". Fortune. 2016-02-29. Retrieved 2016-06-22.
  3. 1 2 Ina Fried (August 9, 2016). "Intel is paying more than $400 million to buy deep-learning startup Nervana Systems". Vox. Retrieved August 22, 2023.
  4. ↑ Jordan Novet (August 9, 2016). "Intel acquires deep learning startup Nervana for more than $350 million". Venture Beat.
  5. 1 2 "Nervana open-sources its deep-learning software, said it would outperform Facebook, Nvidia tools". VentureBeat. 6 May 2015. Retrieved 2016-06-22.
  6. ↑ ""Not so fast, FFT": Winograd - Nervana". Nervana. 2016-03-03. Retrieved 2016-06-22.
  7. ↑ "Nervana's cloud platform makes deep learning more widely available" . Retrieved 2016-06-24.
  8. ↑ Patterson, Steven Max (31 May 2016). "Startup Nervana joins Google in building hardware tailored for neural networks". Network World. Retrieved 2016-06-22.
  9. ↑ Hackett, Robert (November 17, 2017). "Uber's CEO Comes From What May Be the World's Most Techie Family". Fortune .
  10. ↑ "Deep Learning at Scale: Q&A with Naveen Rao, Nervana Systems". re-work.co. Retrieved 2016-06-22.
  11. ↑ "Deep learning startup Nervana raises $20.5M". VentureBeat. 4 June 2015. Retrieved 2016-06-22.
  12. ↑ "Nervana Nevermore: Intel Shifts Focus to Habana Labs, Cancels NNP-T, NNP-I - ExtremeTech". www.extremetech.com. Retrieved 2020-03-30.
This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.