IDC's Worldwide Performance-Intensive Computing Taxonomy, 2022
Description
IDC's Worldwide Performance-Intensive Computing Taxonomy, 2022
This IDC study lays the groundwork for performance-intensive computing (PIC) as a composite market. It combines infrastructure approaches across three major and fast-growing use case categories: high-performance computing (HPC; [aka modeling and simulation [M&S]), artificial intelligence (AI), and Big Data and analytics (BDA). PIC combines infrastructure on and off premises and self-managed and as-a-service deployments. As a compute, storage, and connectivity paradigm, it offers the most powerful and efficient way to execute mathematically intensive, very complex instructions or to perform a relatively simple instruction on massive amounts of data."In the past couple of years, a common infrastructure paradigm has emerged, thanks to the importance of mathematically intensive computations in many use cases found in digital organizations," said Peter Rutten, research VP at IDC's Infrastructure Systems, Platforms, and Technologies Group. "We believe that performance-intensive computing offers an opportunity for vendors to approach these use cases in a more streamlined and targeted fashion."
Please Note: Extended description available upon request.
This IDC study lays the groundwork for performance-intensive computing (PIC) as a composite market. It combines infrastructure approaches across three major and fast-growing use case categories: high-performance computing (HPC; [aka modeling and simulation [M&S]), artificial intelligence (AI), and Big Data and analytics (BDA). PIC combines infrastructure on and off premises and self-managed and as-a-service deployments. As a compute, storage, and connectivity paradigm, it offers the most powerful and efficient way to execute mathematically intensive, very complex instructions or to perform a relatively simple instruction on massive amounts of data."In the past couple of years, a common infrastructure paradigm has emerged, thanks to the importance of mathematically intensive computations in many use cases found in digital organizations," said Peter Rutten, research VP at IDC's Infrastructure Systems, Platforms, and Technologies Group. "We believe that performance-intensive computing offers an opportunity for vendors to approach these use cases in a more streamlined and targeted fashion."
Please Note: Extended description available upon request.
Table of Contents
24 Pages
- IDC's Worldwide Performance-Intensive Computing Taxonomy
- Performance-Intensive Computing Taxonomy Changes for 2022
- Taxonomy Overview
- Note on Massively Parallel Computing
- Advice for the Technology Supplier
- Performance at All Costs
- Package Performance-Intensive Computing Software Stacks
- Low-Latency Interconnect and Network Support
- Definitions
- What Is Performance-Intensive Computing?
- Performance-Intensive Computing Infrastructure
- Infrastructure Hardware
- Discrete Computing Platforms
- General Purpose (COTS Hardware)
- Special Purpose (Custom Hardware)
- Discrete Storage Platforms and Systems
- External Storage Systems
- Server-Based Storage Platforms (Storage-Intensive Servers)
- Converged and Hyperconverged Systems
- Certified Reference Systems and Integrated Infrastructure
- Integrated Platforms
- Hyperconverged Systems
- Connectivity
- Intra-Node Connectivity
- Inter-Node and Datacenter Connectivity
- Wide Area Network Connectivity
- Infrastructure Software
- Physical and Virtual Computing Software
- Operating Systems and Subsystems
- Software-Defined Compute
- Storage Software
- Systems Management Software
- Other Infrastructure Software
- Performance-Intensive Computing Infrastructure Additional Attributes — Hardware and Software (Compute)
- Computing Instance Type
- Core Density Optimized
- Memory Optimized
- Accelerated
- Network Optimized
- Heterogenous
- Parallelization
- Traditional
- Cloud Native
- Scaling
- Level 1: Intra-Node (Processor)
- Level 2: Intra-Node (Accelerator)
- Level 3: Inter-Node (Cluster)
- Level 4: Inter-Node (Datacenter)
- Level 5: Inter-Node (Geodispersed)
- Performance-Intensive Computing Infrastructure Additional Attributes — Hardware and Software (Storage and Connectivity)
- Storage Data Organization
- File Storage Type
- Connectivity Protocols
- Performance-Intensive Computing Infrastructure Additional Attributes — Workloads
- Workload Abstraction
- Bare Metal
- Virtualized
- Containerized
- Workload Profile
- Compute Intensive
- Memory Intensive
- Data Intensive
- Network Intensive
- Hybrid
- Workloads/Use Case Groups
- Artificial Intelligence
- High-Performance Computing
- Big Data and Analytics
- Engineering/Technical
- Other Performance-Intensive Workloads
- Performance-Intensive Computing Infrastructure Additional Attributes — Management and Deployment
- Control Plane
- Premises Based (Installed or as a Service)
- Cloud Based (as a Service)
- Hybrid Cloud and Multicloud Control Planes
- Deployment Location
- Self-Owned and/or Operated Facilities (Local)
- Service Provider Facilities (Hosted)
- Hybrid Deployment
- Management Type
- Self-Managed (Operated)
- OEM Vendor
- Independent Software Vendor
- Cloud Service Provider
- Managed Service Provider
- Performance-Intensive Computing Infrastructure as a Service
- Infrastructure Hardware as a Service
- Infrastructure Software as a Service
- Add-On PaaS and SaaS Options
- Platform-as-a-Service Options
- Software-as-a-Service Options
- Floating Point Precision (Future Consideration)
- Related Markets
- Learn More
- Related Research
- Synopsis
Pricing
Currency Rates
Questions or Comments?
Our team has the ability to search within reports to verify it suits your needs. We can also help maximize your budget by finding sections of reports you can purchase.
