Line

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

oOo

Numascale Whitepapers

NumaConnect

A whitepaper that introduces you to the NumaChip and the technology →download pdf

 

The HTML version of the whitepaper can be found →here

 

Brief:
Numascale's NumaConnect™ technology enables computer system vendors to build scalable servers with the functionality of enterprise mainframes at the cost level of clusters.

 

NumaConnect differentiates from all other interconnects in providing a true high-performance ccNuma architecture and thereby the ability to provide unified access to all resources in a system and utilize caching techniques to obtain very low latency.

 

SMP Redux: You Can Have It All

A Numascale whitepapper by Douglas Eadline→download pdf

 

The HTML version of the whitepaper can be found →here

 

Large scale plug-and-play SMP delivers mainframe performance at commodity prices

 

When High Performance Computing (HPC) is mentioned, one often envisions a large expensive mainframe. These machines are highly parallel and employ a Symmetric Multi-Processing (SMP) design that provides global memory and process spaces. Delivering this level of performance has always been expensive due to the custom engineering required to integrate a large number of processors into a shared memory environment. Indeed, the added expense of large-scale SMP systems has pushed the market to use a “cluster approach,” where a large number of commodity server machines connected together are used as a single resource. It is well known that clusters are harder to manage and less efficient than SMP mainframe systems.

 

In this paper we will contrast current SMP and cluster designs in the context of HPC. We will also introduce a breakthrough technology from Numascale that allows commodity hardware to be combined into a cost-effective and scalable SMP system that delivers the best of both worlds.

 

A Qualitative Discussion of NumaConnect vs InfiniBand and other high-speed networks

A Numascale whitepapper →download pdf

 

The HTML version of the whitepaper can be found →here

 

Brief:

The main focus of InfiniBand was to be able to encapsulate any packet format and provide high bandwidth connections between systems and their peripherals as well as between systems. InfiniBand is a "shared nothing" architecture. This means that all communication requires a driver level in software to control the communication and handle buffers for the RDMA (Remote Direct Memory Access) engines.

 

NumaConnect provides a scalable, low-latency, high bandwidth interconnect with full support for cache coherence. The shared address space allows processors to access remote memory directly without any software driver level intervention and no overheads associated with setting up RDMA transfers. A NumaConnect system to addresses all memory and all memory mapped I/O devices directly.

The Change is on

 

From Cluster

cluster

shared memory with ccnuma by numachip

To Scalable ccNUMA SMP
Cache Coherence - ccNuma - Clusters - Coherent - Directory Based Cache Coherence - Hypertransport - InfiniBand - Numa - NumaChip - Numascale - Snooping
Line