Browse Definitions :

BACKGROUND IMAGE: iSTOCK/GETTY IMAGES

This content is part of the Essential Guide: Keep your server sharp and upgrade memory techniques
Definition

NUMA (non-uniform memory access)

NUMA (non-uniform memory access) is a method of configuring a cluster of microprocessor in a multiprocessing system so that they can share memory locally, improving performance and the ability of the system to be expanded. NUMA is used in a symmetric multiprocessing ( SMP ) system. An SMP system is a "tightly-coupled," "share everything" system in which multiple processors working under a single operating system access each other's memory over a common bus or "interconnect" path. Ordinarily, a limitation of SMP is that as microprocessors are added, the shared bus or data path get overloaded and becomes a performance bottleneck. NUMA adds an intermediate level of memory shared among a few microprocessors so that all data accesses don't have to travel on the main bus.

NUMA can be thought of as a "cluster in a box." The cluster typically consists of four microprocessors (for example, four Pentium microprocessors) interconnected on a local bus (for example, a Peripheral Component Interconnect bus) to a shared memory (called an "L3 cache ") on a single motherboard (it could also probably be referred to as a card ). This unit can be added to similar units to form a symmetric multiprocessing system in which a common SMP bus interconnects all of the clusters. Such a system typically contains from 16 to 256 microprocessors. To an application program running in an SMP system, all the individual processor memories look like a single memory.

When a processor looks for data at a certain memory address, it first looks in the L1 cache on the microprocessor itself, then on a somewhat larger L1 and L2 cache chip nearby, and then on a third level of cache that the NUMA configuration provides before seeking the data in the "remote memory" located near the other microprocessors. Each of these clusters is viewed by NUMA as a "node" in the interconnection network. NUMA maintains a hierarchical view of the data on all the nodes.

Data is moved on the bus between the clusters of a NUMA SMP system using scalable coherent interface (SCI) technology. SCI coordinates what is called "cache coherence" or consistency across the nodes of the multiple clusters.

SMP and NUMA systems are typically used for applications such as data mining and decision support system in which processing can be parceled out to a number of processors that collectively work on a common database. Sequent, Data General, and NCR are among companies that produce NUMA SMP systems.

This was last updated in March 2011

Dig Deeper on Hardware

Join the conversation

1 comment

Send me notifications when other members comment.

Please create a username to comment.

Hi, 

I'm reading on the topic in the 5th edition of "Computer architecture : a quantitative approach" (Hennessy, John L ; Patterson, David A). 

In Ch.5. the authors write that "SMP architectures [...] are called uniform memory access (UMA) multiprocessors, since all processors have a uniform latency from memory, even if the memory is organized into multiple banks.".

And later on "A distributed shared memory (DSM) multiprocessor is also called a NUMA, since the access time depends on the location of a data word in memory".

This seems to contradict the definitions given in the article. Or am I missing something out?

Best,

Simona

Cancel

-ADS BY GOOGLE

File Extensions and File Formats

Powered by:

SearchCompliance

  • risk management

    Risk management is the process of identifying, assessing and controlling threats to an organization's capital and earnings.

  • compliance as a service (CaaS)

    Compliance as a Service (CaaS) is a cloud service service level agreement (SLA) that specified how a managed service provider (...

  • data protection impact assessment (DPIA)

    A data protection impact assessment (DPIA) is a process designed to help organizations determine how data processing systems, ...

SearchSecurity

  • identity theft

    Identity theft, also known as identity fraud, is a crime in which an imposter obtains key pieces of personally identifiable ...

  • cybercrime

    Cybercrime is any criminal activity that involves a computer, networked device or a network.

  • cybersecurity insurance (cybersecurity liability insurance)

    Cybersecurity insurance, also called cyber liability insurance or cyber insurance, is a contract that an entity can purchase to ...

SearchHealthIT

SearchDisasterRecovery

  • disaster recovery plan (DRP)

    A disaster recovery plan (DRP) is a documented, structured approach that describes how an organization can quickly resume work ...

  • business continuity plan (BCP)

    A business continuity plan (BCP) is a document that consists of the critical information an organization needs to continue ...

  • disaster recovery team

    A disaster recovery team is a group of individuals focused on planning, implementing, maintaining, auditing and testing an ...

SearchStorage

  • logical unit number (LUN)

    A logical unit number (LUN) is a unique identifier for designating an individual or collection of physical or virtual storage ...

  • NVMe over Fabrics (NVMe-oF)

    NVMe over Fabrics, also known as NVMe-oF and non-volatile memory express over fabrics, is a protocol specification designed to ...

  • CIFS (Common Internet File System)

    CIFS (Common Internet File System) is a protocol that gained popularity around the year 2000, as vendors worked to establish an ...

Close