HPE ProLiant servers with Xilinx Alveo provide live video transcoding with revolutionary performance that paves a path forward for companies looking to improve TCO.
Production, post production, archiving, and distribution are like islands, each operating based on its own requirements, standards, and practices. Transcoding bridges the gaps between them, enabling media exchange between systems and enabling the vast range of use cases in play today.
What video service providers need now more than ever
Video service providers are faced with balancing high-quality experiences for their customers while managing infrastructure and internet bandwidth operating costs. Given the computational intensity of converting video, transcoding has prompted the need for adaptable hardware acceleration.
Today, video service providers manage bandwidth while guaranteeing high-quality video distribution by utilizing next-generation compression standards and Adaptive Bitrate (ABR) streaming protocols. Both approaches reduce bitrates but increase computational complexity.
To meet these needs, HPE ProLiant and Xilinx Alveo have collaborated to build a solution that provides revolutionary performance for the best user experience.
Pain points for video service providers solved by HPE ProLiant and Xilinx Alveo
Successful video transcoding based on HPE ProLiant servers with Xilinx Alveo helps business customers across all verticals realize significant TCO savings by:
- Offloading server processor workloads to Xilinx Alveo accelerator cards
- Lowering bitrates to generate bandwidth cost savings
- Reducing power consumption (due to fewer required servers) with the ability to utilize an optimized, domain-specific compute architecture for the workload
The joint solution also helps service providers to:
- Deploy one platform and take advantage of future lower bitrate video standards with minor software changes
- Develop software seamlessly using industry-standard FFmpeg commands and enabling custom-framework integration as needed
- Easily scale up and out to handle hundreds of streams and manage clusters of servers with VMware Virtual Machines, or Kubernetes and Docker containers
HPE ProLiant features strengthen this solution
HPE ProLiant servers offer 360-degree security, manageability through intelligent automation, and overall performance optimization of server infrastructure, plus programmable offloading provided by Alveo cards.
360-degree security
Security is at the top of mind for any implementation of a virtualization environment. The enhanced holistic 360-degree security with the silicon root of trust that comes with the HPE ProLiant ecosystem helps assure that security concerns are being met.
Intelligent automation
Migrating new end-user resources into the data center means ensuring that the platforms that host those resources are highly manageable. This a key element of HPE ProLiant’s intelligent automation capabilities.
Performance optimization
The end user’s virtualized experience must be virtually identical to a local experience in order for the implementation to succeed. This means the underlying server platforms must be optimized to meet the critical performance requirements of the solution. The workload optimization capabilities of ProLiant servers ensure the maximum performance of the virtualization experience.
HPE ProLiant solution components
The HPE ProLiant DL385 server is a secure, resilient server that delivers world-class performance and versatility. This is an optimal solution for power users that takes advantage of Xilinx’s felxibility and offloading capability.
Xilinx Alveo optimized for video transcoding
Xilinx Alveo provides real-time real time computing video appliances for easy-to-scale, ultra-high-density video transcoding applications. The solution is ideal for service providers delivering applications such as eSports and game streaming platforms, social and video conferencing, live distance learning, telemedicine, and live broadcast video. It features optimized video quality and bitrate at the lowest cost per channel for significant TCO savings over both software-based and fixed-architecture approaches.
Designed for edge and on-premises compute-intensive workloads where video channel density, throughput, and latency are critical requirements, the new Xilinx Real-Time Video Appliances feature optimized hardware architectures and software to deliver the industry’s highest channel density and lowest latency performance.
A closer look at why this live video transcoding solution is so compelling
Historically, using server-based, CPU-only encoding was possible when most of the video was standard definition (SD) and it made up a small percentage of the overall network traffic. Now that videos consume 80% of network bandwidth and resolutions continue to increase to high definition (HD( and 4K, better compression is needed to keep the bitrates as low as possible. However, the codec complexity is 1000x higher, and CPUs are not efficient at encoding and struggle with this 1000x increase.
That’s why a new specialized compute accelerator is required to handle the job of encoding and processing live real-time video streaming. Xilinx Field-Programmable Gate Arrays (FPGAs) are inherently good at accelerating video transcoding because of two key traits. Why? First, processing video pixels are computationally intensive and FPGA architecture is well suited to this computation. Second, the same Xilinx accelerator board can be updated with software updates to improve and optimize existing codecs as well as new codecs as they evolve. Unlike ASIC-based products, this approach future-proofs infrastructure from constant codec evolution.
Xilinx has built the Alveo accelerator platform to put its latest FPGA architecture into HPE ProLiant servers along with an FFmpeg based software stack to significantly increase transcoding throughput and dramatically improve the overall TCO. It’s a winning solution to continue to fuel the accelerating digital media revolution.
Learn more
Mark Azadpour
Hewlett Packard Enterprise
twitter.com/HPE_Servers
linkedin.com/showcase/hpe-servers-and-systems/
hpe.com/servers
Mark_Azadpour
Mark is a senior product manager in workload-solutions team focusing on offload acceleration, virtualization, and VDI. His charter is to bring solutions to market with optimal TCO to enhance customer experience. He has several decade of experience in semiconductor, telecom, HPC, and storage. He has a PhD. in computer engineering, MBA, and PMP.
