Intel disclosed several significant features of its upcoming “Crystal Forest” next-generation communications platform. Intel said Crystal Forest will enable equipment manufacturers to consolidate three communications workloads – application, control and packet processing – on multi-core Intel architecture processors to deliver better performance and accelerate time to market.
Some key points on Crystal Forest:
- Crystal Forest is expected to deliver up to 160 million packets per second performance for Layer 3 packet forwarding.
- Crystal Forest will utilize Intel QuickAssist technology, which processes and accelerates specialized packet workloads – cryptography, compression and deep packet inspection included – on standard Intel platforms. QuickAssist enables secure Internet transactions can be accelerated up to 100 Gbps on the platform.
- A Data Plane Development Kit, which consists of a set of software libraries and algorithms, can improve the performance and throughput of packets on Intel architecture platforms to yield more than five times the performance over previous generations of Intel platforms.
- Additionally, Crystal Forest will use a common application programming interface and common drivers so that multiple designs can be implemented in much less time and at much lower development costs.
- The new platform is scheduled to be available later in 2012.
Intel said its roadmap calls for annual performance refreshes for several years, so equipment manufacturers and service providers will be able to scale and refresh their designs to meet future network needs.
“The demand for increased network performance will continue to grow as more smart devices connect to the Internet every day,�? said Rose Schooler, general manager of Intel’s Communications Infrastructure Division. “And with the popularity of social networking and other high-bandwidth services, such as video and photo uploads/downloads, interactive video, crowdcasting and online gaming, service providers will be challenged to efficiently provision sufficient upstream capacity and manage the spike in networktraffic.�?
