The Decline of Moore’s Law and the Rise of the Hardware Accelerator

Pliops Extends Performance Benefits Tech Industry Has Come to Expect Over Last 55 Years

SAN JOSE, Calif., April 22, 2020 (GLOBE NEWSWIRE) -- In 1965, the world’s first commercial minicomputer was introduced, the first human walked in space and a prediction was made that would describe an incredible technological and economic force. The semiconductor industry made this prediction a benchmark to plan strategies and set research and development targets for the next 55 years. In spring of that year – April 19, to be exact – Intel’s Gordon E. Moore described an exponential growth trend in computing power where the speed and capability of computers could be expected to double every two years.

A Call to Action

Moore's Law created an industry expectation for increasing performance – but all good things must come to an end. 2020 finds Moore's Law dramatically slowing, with processor core performance now forecasted to double every 20 years. However, the cloud data explosion, further accelerated by AI applications, depends upon performance increases in processors. Compounding these concerns, cloud service providers are currently facing more challenges than ever before as they struggle to keep up with the unprecedented traffic being created by the world’s accelerated shift online caused by social distancing and the COVID-19 crisis. Clearly, a new approach to the data center infrastructure challenges of an increasingly dynamic and online world is needed – one that picks up where Moore’s Law is leaving off.

Technology innovator Pliops was founded to address this gap in performance, and has developed a storage processor that enables data centers to operate faster and more efficiently. The Pliops architecture overcomes the major inefficiencies of software-only storage engines via a dedicated hardware product. 

The Case for Specialized Processors

Without a better technology solution, today’s companies have no choice but to throw more and more CPUs and SSDs into the mix in an attempt to address cost-effective performance and scalability. This has led to server sprawl and increased operating expenses. Infrastructure architects and engineering leaders seeking to keep the pedal to the metal in terms of compute performance are starting to look to specialized solutions that accelerate the pace of workflows.

From GPUs to sound cards, the creation of dedicated processors for high-order tasks is not a new concept. In fact, after first appearing on the scene in 2017, artificial intelligence-optimized processors have become the key to faster, more power-efficient machine learning tasks. Shipments of these processors have been forecasted to leap from 340 million last year to 1.5 billion in 2023.

By creating silicon-based accelerators for cloud databases and infrastructures, Pliops has taken the same approach employed by AI processors and applied it to more common – but no less challenging – workloads. At the core of the Pliops solution is a powerful FPGA from Xilinx and high-performance hardware algorithms. 

“Hardware acceleration has proven itself as the answer to achieving performance gains in the post Moore’s Law era,” said Freddy Engineer, corporate vice president and general manager, data center group, at Xilinx. “By implementing analytics processing in an adaptable compute platform from Xilinx, Pliops is helping to expand hardware acceleration into data intensive tasks with direct access to storage.”

A New Acceleration Option

“Core operational and transactional databases hold critical data that is vital to a company and data center’s operation, but their software and architectures are so inefficient that architects create a huge protective ‘bubble’ around them,” explained David Nagle, advisor to Pliops. “Data centers that embrace specialized processors to offload common applications can not only reduce server count, storage capacity and network costs, but can also share that hardware across both their transactional and analytic workloads. This is a game-changer in terms of efficiency, performance, scalability, costs, and shared real-time access to data.”

Pliops’ architecture creates the closest possible connection between application and flash storage, eliminating decades of accumulated legacy application, file system and storage layers and the associated inefficiencies.

“Specialized processors are changing the face of computing and allowing the innovative spirit of Moore’s Law to live on,” Nagle continued. “Adopting them does require a shift in thinking, as well as a change in infrastructure, but the benefits gained deliver value well beyond what traditional processors can deliver today and enable scaling for the next decade and beyond. This is especially true when applied to cloud databases, storage systems, and analytics platforms where Pliops is delivering 10x faster performance and 100x improved response time in less than half the amount of storage. The greater the challenge, the greater the reward.”

“On this 55th anniversary we mark the end of only the first ‘semiconductor process’ phase of Moore’s Law,” said Kevin Deierling, senior vice president of marketing for Mellanox Technologies. “Now we are entering a new phase where the march of performance and efficiency will continue, with advances from chip processing slowing, but accelerating contributions from innovation of chip and system architectures and optimization at data center scale. Such innovation includes acceleration and parallelism across the entire data center stack, including accelerated computing from GPUs, networking, storage, and software. Innovations like Pliops’ database acceleration and intelligent networking from Mellanox are critical to keeping this next phase of Moore’s Law marching forward.”

"While the slowdown of Moore's Law has significantly disrupted the industry, it has also created opportunities," added Henri Richard, executive vice president of worldwide field and customer operations for NetApp and a semiconductor industry veteran. "Companies like NetApp and Pliops that recognize Moore's Law for what it truly was – a mandate for innovative thinking – will be the ones to lead the future of computing forward. The decline of Moore's Law should be seen as the dawn of a new age."

Please visit www.pliops.com for more information.

About Pliops
Founded in 2017, Pliops is a technology innovator focused on making data centers run faster and more efficiently. Its technology addresses skyrocketing data volumes and solves the slowing compute performance problem. The company's Storage Processor is built upon a groundbreaking patent-pending approach that accelerates storage functions. Focused on creating the next wave of the accelerated data center, the Pliops Storage Processor enables cloud and enterprise customers to access data up to 100 times faster – using just a fraction of the computational load and power consumption. With Pliops technology, databases, analytics and other data-intensive applications are able to reach their full potential. Investors include Softbank Ventures Asia, Intel Capital, Western Digital, Mellanox, and Xilinx.

For more information on Pliops technology or career opportunities visit  www.pliops.com or connect on  Twitter and  LinkedIn.

Media Contact:
Stephanie Olsen
Lages & Associates
(949) 453-8080
stephanie@lages.com

Primary Logo

Featured Video
Jobs
Currently No Featured Jobs
Upcoming Events
Consumer Electronics Show 2025 - CES 2025 at Las Vegas Convention Center NV - Jan 7 - 10, 2025
ESD Alliance "Savage on Security” Webinar at United States - Jan 23, 2025
SEMICON Korea 2025 at Hall A, B, C, D, E, GrandBallroom, PLATZ, COEX, Seoul Korea (South) - Feb 19 - 21, 2025
DVCon U.S. 2025 at United States - Feb 24 - 27, 2025



© 2024 Internet Business Systems, Inc.
670 Aberdeen Way, Milpitas, CA 95035
+1 (408) 882-6554 — Contact Us, or visit our other sites:
AECCafe - Architectural Design and Engineering TechJobsCafe - Technical Jobs and Resumes GISCafe - Geographical Information Services  MCADCafe - Mechanical Design and Engineering ShareCG - Share Computer Graphic (CG) Animation, 3D Art and 3D Models
  Privacy PolicyAdvertise