Improve Price Performance by Up to 51% with Amazon and Intel OpenSearch.
Users can enjoy notable speed gains in vector search workloads by utilising the capabilities of Intel Advanced Vector Extensions 512 (Intel AVX-512) technology. In particular, the new C/M/R 7i instances can achieve up to 51% better pricing performance than the old R5 Intel instances when OpenSearch 2.17 or newer domains are run on them.
When upgrading to the newest instances, customers that want to maximise their search skills will find the Intel AVX-512, a built-in accelerator, to be extremely helpful. OpenSearch automatically activates these accelerators, so no code or configuration modifications are needed to boost performance.
What is Intel AVX-512?
In high-performance computer applications, Intel AVX-512 instructions accelerate complex activities. The Intel AVX-512 may greatly speed up processes involving huge datasets, including vector searches, which are frequently used in machine learning, data analytics, and scientific computing, by turning on these instructions.
Why Upgrade to C/M/R 7i Instances?
The most recent C/M/R 7i instances are made to take full advantage of Intel AVX-512 technology. These instances are accessible globally through Amazon OpenSearch Service and provide increased processing power, memory bandwidth, and optimised performance for compute-intensive tasks (see availability). Upgrade to take full advantage of OpenSearch’s search features.
Accelerate Your Compute-Intensive Workloads
AVX-512, Intel Advanced Vector Extensions 512. Improved computing performance is desired across all industries. With Intel Advanced Vector Extensions 512 (Intel AVX-512), available on Intel Xeon Scalable processors, it continue to legacy of workload optimized innovation to meet growing needs and changing usage models.
High-Performance Computing
Complex system modelling, engineering simulations, and scientific research all depend on high performance computing, or HPC. Intel acceleration technologies enable you to get the performance you need for these taxing operations.
Learn how 3rd Gen Intel Xeon Scalable processors’ built-in acceleration from the Intel AVX-512 instruction set will help you tackle your most difficult computing tasks.
Performance Boost for Demanding Workloads
A new set of instructions called Intel Advanced Vector Extensions 512 (Intel AVX-512) can speed up performance for tasks and applications like financial analytics, 3D modelling and analysis, image and audio/video processing, scientific simulations, deep learning, cryptography, and data compression.
Improved Vector Processing Skills
Your most demanding computing jobs can be handled by the Intel AVX-512 to its ultra-wide 512-bit vector operations capabilities.
With up to two 512-bit fused-multiply add (FMA) units, applications can pack 32 double precision and 64 single precision floating point operations per clock cycle within the 512-bit vectors. This doubles the width of data registers, doubles the number of registers, and doubles the width of FMA units in comparison to Intel Advanced Vector Extensions 2 (Intel AVX2).2 3
Related Products and Solutions
Accelerate Performance with Intel AVX-512
The newest x86 vector instruction set, Intel Advanced Vector Extensions 512 (Intel AVX-512), has up to two fused-multiply add units and additional optimizations to boost performance for the most taxing computing jobs.
Server Items
Intel processors power every task and increase data centre productivity, from memory-intensive business data to the cloud.
Intel Xeon Processor Family
Provide big data insights, cloud computing, real-time analytics, and processing for your vital company operations. Boost data centre dependability and efficiency to manage any workload.
Enterprise Data Center
Gain access to materials based on Intel’s extensive knowledge of data centres.
Comprehending Intel Crypto Acceleration
To comprehend and maximize your usage of these instructions, which are now available on third-generation Intel Xeon Scalable processors, learn more about Intel AVX-512 and Intel Crypto Acceleration.
The computationally demanding indexing and search procedures necessitate computations between vectors, which are commonly represented as 128–2,048 dimensions (numerical values). The Intel AVX-512 is used by the Intel Xeon Scalable processors on 7th generation Intel instances to boost vector operations’ speed and effectiveness with the following features:
- Parallel processing of data: The Intel AVX-512 effectively employs SIMD (single input multiple data) to conduct several operations simultaneously, providing a considerable speedup, by processing 512 bits of data at once, which is twice as much as its predecessor.
- Pathlength reduction: A notable improvement in pathlength a metric that quantifies the amount of instructions needed to complete a unit of work in workloads is the cause of the speedup.
- Power performance savings: By processing more data and carrying out more processes in less time, you can reduce power performance expenses.
