Enterprise Technology Terminology

5,405 results · page 231 of 271

• Sparse Tensor Engine

  Sparse tensor engine is a hardware or software execution unit for tensor algebra on sparse data that skips zero elements, enabling more efficient AI, analytics, and scientific workloads for enterprises within power, cost, and capacity constraints.

• Sparse Transformer

  Sparse Transformer is a transformer neural network variant that uses sparse self-attention patterns to reduce computation and memory requirements for long sequences, which supports enterprise workloads that need extended context processing within constrained training and inference infrastructure budgets.

• Spatial AI Processor

  Spatial AI processor is a specialized compute architecture that runs machine learning and computer vision workloads on three-dimensional spatial data at the edge, enabling on-device perception, mapping, and scene understanding for robots, vehicles, cameras, and augmented or mixed reality systems in enterprise environments.

• Spatial Computing

  Spatial computing integrates digital content, computation, and user interaction with three-dimensional physical environments using AR, VR, MR, sensor, and mapping technologies, and matters in enterprise contexts for contextual workflows, digital twins, remote collaboration, and integration with existing operational and data platforms.

• Spatial Multiplexing

  Spatial multiplexing is a MIMO transmission technique that sends parallel data streams over separate spatial paths in the same time and frequency resources, enabling higher spectral efficiency and capacity in enterprise Wi-Fi, LTE, and 5G network deployments.

• Spatial Reuse

  Spatial reuse is a wireless networking technique that reuses the same time and frequency resources in different physical locations under controlled interference, enabling higher capacity and spectrum efficiency in dense Wi-Fi and cellular deployments for enterprise and carrier environments.

• Spatial Simulation Model

  Spatial simulation model is a computational representation of spatially distributed processes that simulates how entities or conditions evolve over time across geographic or geometric space. It matters for enterprises that need location-dependent scenario analysis for planning, risk assessment, and resource allocation.

• Speach Models

  Speech models are machine learning models that analyze and generate human speech audio for tasks such as recognition, transcription, speaker characterization, and synthesis, and they matter in enterprise environments because they convert voice data into structured, searchable, and automatable digital assets.

• Specific Language Models

  Specific language models are language models trained or adapted on constrained, domain-specific or organization-specific data and vocabularies, used by enterprises to produce context-aligned outputs that match internal terminology, policies, and regulatory requirements better than general-purpose models in targeted applications.

• Spectrum

  Spectrum is the continuous range of electromagnetic frequencies and wavelengths that wireless and optical communication systems use for transmission and sensing, and it matters because enterprise connectivity, private networks, and wireless architectures depend on access to specific spectrum bands under regulatory constraints.

• Spectrum Allocation

  Spectrum allocation is the regulatory process that designates specific radio frequency bands for defined wireless services and users, enabling orderly spectrum use, interference control, and predictable access for mobile, satellite, Wi-Fi, private networks, and other enterprise communication systems across jurisdictions.

• SPHINCS+

  SPHINCS+ is a stateless hash-based post-quantum digital signature scheme standardized by NIST that uses only cryptographic hash functions, providing a quantum-resistant option for integrity, authentication, and nonrepudiation in enterprise systems, public key infrastructures, and long-term data protection strategies.

• Spike-Timing Dependent Plasticity

  Spike-timing dependent plasticity (STDP) is a timing-based synaptic learning rule where the order and interval of neuronal spikes alter synaptic strength. STDP matters in enterprise contexts as a foundation for neuromorphic computing, spiking neural networks, and low-power on-device learning architectures.

• Spiking Encoder

  Spiking encoder is a component in spiking neural network and neuromorphic systems that converts sensor or feature inputs into spike trains, enabling spike-based computation and affecting accuracy, latency, and energy properties of enterprise AI and edge workloads.

• Spiking Neural Network

  Spiking neural network is a class of artificial neural network that uses discrete spikes and temporal coding for computation, relevant to enterprises evaluating low-power, event-driven AI methods for edge inference, temporal data processing, and neuromorphic hardware deployments.

• Spine-Leaf Architecture

  Spine-leaf architecture is a two-tier data center network design that uses spine and leaf switches in a full-mesh fabric to provide predictable latency, uniform bandwidth between servers, and scalable connectivity for virtualized, containerized, and multi-tenant enterprise environments.

• Spine Switch

  Spine switch is a high-capacity data center network device deployed in the spine layer of a leaf-spine architecture, providing uniform, low-latency connectivity between leaf switches and enabling scalable east-west traffic handling for virtualized, cloud, and distributed enterprise workloads.

• Spin Qubit

  Spin qubit is a type of qubit that stores quantum information in the spin state of an electron or nucleus, relevant to enterprises as a candidate hardware platform for scalable quantum processors built on semiconductor and solid-state technologies.

• Spot Beam

  Spot beam is a focused satellite antenna beam that covers a limited geographic area, enabling higher power, greater spectral efficiency, and frequency reuse. It matters in enterprise satellite networking for planning coverage, capacity, and service quality across specific regions or routes.

• Spot Instance Management

  Spot instance management is the governance, automation, and operational control of interruptible, discounted cloud compute capacity purchased from excess provider resources, used by enterprises to reduce compute spending while keeping defined limits on interruption risk, workload placement, and service-level objectives.