UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

SC18: Adaptive acceleration, deployment-ready solutions for your data center.

Booth Highlights

Booth Demonstrations

Inference Acceleration with IBM PowerAI Vision
This demonstration shows computer vision made simple. Learn how you can train and deploy models on IBM Power Systems while integrating Xilinx ML Suite for inference.

Inference Acceleration in High Density Compute Server
This demonstration shows a dense AMD EPYC server supporting up to eight Alveo U200/U250 accelerator cards allowing for the highest adaptable compute density for compute intensive applications. The system also features a motherboard from Supermicro.  This system can be used to accelerate applications in the areas of genomics, video, big data, finance and more.

Big Data Analytics with Computational Storage
This demo shows BigStream Apache-Spark running 4X faster with Samsung Smart SSDs. The Samsung Smart SSD utilizes the Xilinx ZU19EG MPSoC device for computational storage acceleration resulting in faster time to insight, fewer CPU resources, and the ability to add FPGA-Compute + SSD capacity in granular increments independent of the CPU resources.

NVMe-attached Accelerated Compression
This demonstration showcases over 100Gbps of data compression using Eideticom NVMe-attached NoLoad™ acceleration. The NoLoad 250-U2 FPGA Accelerator uses an NVMe-compliant interface and supports Peer to Peer (P2P) transfers of PCIe data via Controller Memory Buffers (CMBs).The HPE DL385 server with AMD EPYC processor enables P2P PCIe data transfers to/from a NoLoad 250-U2 Xilinx-enabled FPGA card. NoLoad storage and analytics accelerator performs compute functions like compression and processes/moves the data to adjacent SSD’s with minimal host CPU intervention.

Hadoop Ingest with Computational Storage
Modern Hadoop cluster workloads are scaling into the Petabyte range, stressing infrastructure from both compute resource and storage I/O perspectives. ScaleFlux Computational Storage is the first turnkey, easy-to-deploy solution that accelerates GZIP compression and Erasure Coding processing (Hadoop 3), the main bottlenecks for big data ingest. In addition, ScaleFlux opens storage I/O by adding low-latency PCIe Flash as temporary storage space for analytical processing jobs assigned to HDDs. The end result? Optimized time-to-insight from massive data workloads. This demo utilizes standard Hadoop Teragen and Terasort benchmarks to show how ScaleFlux Computational Storage can cost-effectively deliver 3x faster data ingest times while running 60% more jobs on the same infrastructure footprint.

Key-Value Store Acceleration with CCIX
This demonstration shows the seamless KVS acceleration enabled by CCIX. See how you can accelerate KVS while maintaining the network inference solution and intelligence in application control operations processing. CCIX enables shared data structures between host and accelerator processing threads while maintaining high throughput with Storage Class Memory for large capacity solutions.

There will be more exciting announcements and demos at the show, come see us at booth #927.

Learn more about Alveo.

Booth Information

Xilinx Booth #927
Kay Baily Hutchison Convention Center
Dallas, TX

Monday, November 12
7:00 PM – 9:00 PM

Tuesday, November 13
10:00 AM – 6:00 PM

Wednesday, November 14
10:00 AM – 6:00 PM

Thursday, November 15
10:00 AM – 3:00 PM

Learn More

Conference Participation

Conference Participation

IA3 2018 8th Workshop on Irregular Applications: Architectures and Algorithms
Monday, November 12 | 2:10 PM – 3:00 PM
Room D172,  Kay Bailey Convention Center

Keynote | Versal: The new Xilinx Adaptive Compute Acceleration Platforms (ACAP)
Presented by Kees Vissers, Distinguished Engineer, Xilinx
An overview of the new Adaptive Compute Acceleration Platform (ACAP) including the system architecture of the family of devices. This presentation will focus on the new AI engines in detail and show the concepts for the programming environment, the architecture, the integration in the total device, and some application domains including Machine Learning and 5G wireless applications.

Partner’s Theater

Joint Presentation with Inspur
Wednesday, November 14 at 11:00 AM | Inspur Booth # 1235
Presented by Freddy Engineer, Vice President, Global Data Center Sales, Xilinx

FPGA Acceleration in the Data Center
Thursday, November 15 at 11:30AM | Dell Booth #3218
Presented by Jim Dworkin, Director of Data Center Marketing, Xilinx

Ecosystem Partners

Ecosystem Partners

Visit our ecosystem partners on the show floor at SC18 to see demonstrations of Xilinx technology.  

Company Name Booth #
Algo-Logic 3854
Alpha Data 222
AMD 2824
Arista / Metamako 2022
BittWare 226
CCIX Consortium 4411
Dell 3218
Gen-Z Consortium 4101
LDA 611
Liqid 3841
Inspur 1235
IntelliProp Inc. 4145
Mellanox 3207
Micron 3247
Mipsology 3871
Molex 3941
Nimbix 1600
OpenCAPI Consortium 627
Page Bookmarked