TFLOPS


Also found in: Dictionary, Encyclopedia.
Related to TFLOPS: teraflop, Petaflop, gigaflop
AcronymDefinition
TFLOPSTera (trillion) Floating Point Operations per Second
References in periodicals archive ?
Cray already delivered an XT4 system to NERSC in January 2007, which will contain more than 19,000 processor cores when fully installed, and a theoretical peak speed of more than 100 TFLOPS.
Intel's "Knights Corner" product, the first commercial co-processor based on the Intel[R] Many Integrated Core (Intel[R] MIC) architecture, was shown for the first time breaking the barrier of 1 TFLOPS double precision performance**.
9 TFLOPS of single precision compute performance while using just 75 watts of power.
When combined with existing SGI equipment, the Altix ICE installation at CINES now provides 267 TFLOPS and 750TB of attached storage.
Supermicro also debuts the density optimized 4U SuperServer 4028GR-TR(T)2 system supporting up to 10 PCI-E Tesla P100 accelerators for up to 187 TFLOPS FP16 peak performance with GPU Direct RDMA support.
Although Sun is currently nowhere to be seen on the higher reaches of the list, it says it is now "seriously competing" for the 30 TFLOPS Accelerated Strategic Computing Initiative Turquoise project, having won an ASCI PathForward award from the US Department of Energy.
This 3 slot slice achieves performance well into the TFLOPS range of processing, and can be replicated multiple times in a chassis to create unprecedented computational performance in an ISR subsystem.
8 TFLOPS performance in a sub-95W TDP power profile, the new embedded GPUs offer advanced 4K encode/decode multimedia processing capabilities and support for up to six displays on the E9550.
84 TFLOPS (Tera-Floating-point Operations Per Second).
Presently the world's fastest vector supercomputer, the SX-9 closes in on the PFLOPS[2] range by realizing a processing performance of 839 TFLOPS.
Both IVM and SGI are working on next generation 10, 30 and 100 TFLOPS, machines under the ASCI program that runs through 2004 (CI No 3,350).
64 TFLOPS of compute power, the new AMD FireStream accelerators are ideal for HPC, cloud and enterprise-scale deployments that require advanced performance for handling today's highly parallel, compute-intensive workloads.