
Embracing the shift to media workflow virtualization, and enabling hardware acceleration of software video processing algorithms, Alveo adaptable accelerator cards provide an ideal platform for development and deployment of a wide range of broadcast applications.
Highest Performance
- Up to 90X higher performance than CPUs on key workloads at one-third the cost
- Over 4X higher inference throughput and 3X latency advantage over GPU-based solutions
Adaptable to Any Workload
- Machine learning inference to video processing/encoding/transcoding to any workload using the same accelerator card
- As workload algorithms evolve, use reconfigurable hardware to adapt faster than fixed-function accelerator card product cycles
Cloud On-Premise Mobility
- Deploy solutions in the cloud or on-premise interchangeably, scalable to application requirements
- Applications available for common workloads, or build your own with the Application Developer Tool
LET US CONNECT YOU TO
Other products from this company:

Alveo Adaptable Accelerator Cards

Kintex UltraScale+ FPGA Family

Spartan-7 FPGA Family

Zynq UltraScale+ MPSoC EV

Zynq UltraScale+ MPSoC EV

Xilinx Real-Time Video Servers
Alveo Adaptable Accelerator Cards
Highest Performance
- Up to 90X higher performance than CPUs on key workloads at one-third the cost
- Over 4X higher inference throughput and 3X latency advantage over GPU-based solutions
Adaptable to Any Workload
- Machine learning inference to video processing/encoding/transcoding to any workload using the same accelerator card
- As workload algorithms evolve, use reconfigurable hardware to adapt faster than fixed-function accelerator card product cycles
Cloud On-Premise Mobility
- Deploy solutions in the cloud or on-premise interchangeably, scalable to application requirements
- Applications available for common workloads, or build your own with the Application Developer Tool
Kintex UltraScale+ FPGA Family
- Up to 1.2M System Logic Cells
- UltraRAM for on-chip memory integration
- Integrated 100G Ethernet MAC with RS-FEC
- 6.3 TeraMACs of DSP compute performance
- Up to 28Gbps transceivers capable of 12G-SDI and 10GE/25GE
- VCXO and fractional PLL integration reduces external clocking component cost
- Xilinx partner IP available for ST 2110 and NMOS
Spartan-7 FPGA Family
These devices feature a MicroBlaze soft processor running over 200 DMIPs with 800Mb/s DDR3 support built on 28nm technology. Additionally, Spartan-7 devices offer an integrated ADC, dedicated security features and highly parallel DSP capabilities making them ideal for both audio and wireless transmission processing with low latency.
- High pin-count to logic ratio for I/O connectivity
- MicroBlaze™ processor soft IP
- Integrated security and monitoring
- Up to 1.25Gb/s LVDS
- 25.6Gb/s peak DDR3-800 memory bandwidth with flexible, soft memory controller
- XADC and SYSMON for integrating discrete analog and monitoring circuitry
- Cost-optimized for system I/O expansion
- 1.0V core voltage or 0.95V core voltage option
- 50% lower total power than 45nm generation devices
- Enabled by Vivado® HLx Design Suite WebPack™
Zynq UltraScale+ MPSoC EV
The Zynq UltraScale+ MPSoC EV devices are particularly suited to a wide range of broadcast and media applications, from camera encoding to mobile wireless video systems, and contribution IRDs to live streaming encoders:
- Configurable and dynamically adjustable codec
- Integrate with a wide portfolio of Xilinx and partner IP cores such as 12G-SDI, ST 2110, HDMI 2.0, DisplayPort 1.2 etc.
- Up to 4K60 4:2:2 10-bit H.264/H.265 encode + decode
- Multichannel support up to 4K60 (e.g. 2x 1080p60 or 4x 1080p30)
- Ultra-low-latency mode of <35ms glass-to-glass
- Total device power for a single pipeline as low as 8.5W
Zynq UltraScale+ MPSoC EV
The Zynq UltraScale+ MPSoC EV devices are particularly suited to a wide range of broadcast and media applications, from camera encoding to mobile wireless video systems, and contribution IRDs to live streaming encoders:
- Configurable and dynamically adjustable codec
- Integrate with a wide portfolio of Xilinx and partner IP cores such as 12G-SDI, ST 2110, HDMI 2.0, DisplayPort 1.2 etc.
- Up to 4K60 4:2:2 10-bit H.264/H.265 encode + decode
- Multichannel support up to 4K60 (e.g. 2x 1080p60 or 4x 1080p30)
- Ultra-low-latency mode of <35ms glass-to-glass
- Total device power for a single pipeline as low as 8.5W
Xilinx Real-Time Video Servers
Introducing the Xilinx Real-Time Video Server appliance reference architectures. Powered by Alveo™ accelerator cards, Xilinx delivers breakthrough live video streaming performance at the lowest cost per channel for significant TCO savings over fixed-architecture approaches, without altering existing infrastructure.
- Up to 5X Faster Than CPUs
- 6X Lower Cost
- Up to 5X Lower Power
Bitrate Optimized Appliance (Alveo U50)
For broadcasters delivering a small number of streams to large viewerships that demand high-quality video transcoding. For live streaming video service providers, OEMs, and Content Delivery Networks, the Alveo bitrate-optimized, and live transcoding solution enables service providers to reduce bandwidth costs (OPEX) by replacing less efficient software-based (CPU) transcoders.
High Channel Density Appliance (Alveo U30)
Optimized for a high volume of broadcasters with a smaller number of viewers demands high-density transcoding. For video service providers faced with the challenge of managing infrastructure costs while simultaneously improving customer quality of experience (QoE), the Alveo high-density video transcoding solution provides optimized cost per channels (reduce CAPEX).
Related Products:

Alveo Adaptable Accelerator Cards

8 Bay / 16 Bay EVO...

nxtedition

Via24

Bare Metal Cloud (BMC)
Alveo Adaptable Accelerator Cards
Highest Performance
- Up to 90X higher performance than CPUs on key workloads at one-third the cost
- Over 4X higher inference throughput and 3X latency advantage over GPU-based solutions
Adaptable to Any Workload
- Machine learning inference to video processing/encoding/transcoding to any workload using the same accelerator card
- As workload algorithms evolve, use reconfigurable hardware to adapt faster than fixed-function accelerator card product cycles
Cloud On-Premise Mobility
- Deploy solutions in the cloud or on-premise interchangeably, scalable to application requirements
- Applications available for common workloads, or build your own with the Application Developer Tool
8 Bay / 16 Bay EVO Shared Storage
- It’s incredibly fast, easily expandable, and its flexibility is unmatched.
- Workflow Project sharing, bin sharing, file sharing, SAN & NAS for professional media applications.
- Connectivity Enough ports to directly connect every computer in your workgroup.
- Performance High throughput, with overhead for the most demanding projects and 6K video workloads.
- Expansion Ready for more capacity and direct-connect network ports as your business continues to grow.
With EVO, you can connect and share everything — you’ll spend less and still get better performance. EVO is purpose-built for digital media workflows and demanding use with the leading pro apps like Avid Media Composer, Final Cut Pro/FCP X, Adobe Premiere Pro/CC, and DaVinci Resolve. When it comes to projectsharing, you can use Avid’s native Project and Bin sharing functions, and take advantage of EVO’s unique project sharing features for FCPX and Adobe Premiere Pro. EVO unlocks the sharing capabilities of Media Composer — with several editors in the same Avid project, teams can easily collaborate by using shared Bins, exchanging access to Media and Sequences directly within Media Composer. You also can Enjoy the flexibility and performance of both file and volume sharing. EVO enables you to use file sharing, volume sharing, or a combination of the two — providing multi-user write workspaces for sharing projects and bins, in addition to high performance SAN volumes for multi-cam and uncompressed workflows.
EVO is designed and built to make your production life easier, which means you get purpose built features like Slingshot and ShareBrowser. With Slingshot, an Automations GUI and API, you can create and run automated file workflows that span EVO, other storage systems, and Amazon S3/Glacier cloud services. Slingshot helps boost productivity by enabling you to automate repetitive file management tasks, leaving you and your team with more valuable time in post. ShareBrowser is an Easy-to-Use, unified file/project/asset management interface for macOS and Windows. It is similar to Finder/Explorer, but designed by SNS specifically to help media teams work and collaborate more efficiently. The ShareBrowser Desktop Client not only improves visibility and access to your team’s storage systems, it conveniently enables automatic project locking for FCP, Premiere Pro, and Pro Tools — as well as Avid project and bin sharing. enables your team to search across all EVO storage as well as local, offline, and other network disks. Assets and custom metadata like tags and comments can be instantly added to anything in the shared database. This means your team can more easily find, preview, and share important media and projects - and now, it's also available in a free web-app to make collaboration even easier.
nxtedition
Via24
Where’s the best place to open a new facility?
In the current landscape, the answer is: wherever it’s needed.
Remote working is becoming more important. Editing and grading on location is commonplace. Major projects are shuttled between multiple continents depending on what stage of post they’ve reached. All this means your team need access to resources from more locations than ever before, in more locations than ever before, making cloud applications and storage critical to your success.
Our Via24 cloud services are designed to give your team access to the cloud applications, storage and high speed file transfers that you need. Our experienced networking and infrastructure team can ensure your media is where it needs to be, delivered securely and managed effectively within your storage hierarchy. Whatever you need to achieve, we can help design and support a sustainable cloud workflow that keeps you ahead of the competition, regardless of location.
Bare Metal Cloud (BMC)
- Dedicated
- Dedicated servers reduce latency by up to 40% vs. virtual machines
- Eliminate “noisy neighbor” effect
- Interconnected
- Connect all 60+ bare metal PoPs via Zenlayer’s private backbone
- Directly connect to public clouds such as AWS, Azure, Google Cloud, Alibaba Cloud, etc.
- On demand
- Activate in 10 minutes
- Add bandwidth/IPs as needed