HAQM EC2 I7i Instances

Best Performance, Highest Density Storage Optimized Instances

Get started with I7i Instances

Why HAQM EC2 I7i Instances?

HAQM EC2 I7i and I7ie instances, powered by 3rd generation AWS Nitro SSDs and 5th generation Intel Xeon Scalable processors with an all-core turbo frequency of 3.2 GHz, offer significant performance improvements for demanding workloads. I7i instances deliver the best compute and storage performance among x86-based HAQM EC2 storage instances, with 23% improved compute performance and more than 10% better price performance compared to previous generation I4i instances. I7ie instances deliver up to 40% improved compute performance and up to 20% better price performance over previous generation I3en instances. Both instance families leverage PCIe Gen5-based AWS Nitro-SSDs, with I7i instances providing up to 45TB of NVMe storage and delivering up to 50% better real-time storage performance, 50% lower storage I/O latency, and 60% reduced I/O latency variability compared to I4i instances. I7ie instances deliver the highest local NVMe storage density in the cloud with up to 120TB of NVMe storage, while delivering up to 65% better real-time storage performance, 50% lower storage I/O latency, and 65% reduced I/O latency variability compared to I3en instances.

I7i and I7ie instances are optimized for storage-intensive workloads that require high-speed access to data residing on local NVMe storage, at real-time latency. The combination of high throughput and low-latency access makes these instances ideal for applications with strict service level agreements (SLAs), including relational databases like MySQL, Oracle DB, and Microsoft SQL Server, as well as NoSQL databases such as MongoDB, Cassandra, Aerospike, Redis, SycllaDB, Clickhouse, RocksDB. This robust architecture ensures reliable performance for data-intensive applications including real-time analytics (Spark, Kafka, Splunk, Databricks), AI LLM pre-processing for training, distributed file systems, search engines, indexing, and blockchain analysis workloads.

Benefits

Built on 3rd generation AWS Nitro SSDs, I7i and I7ie instances deliver industry-leading NVMe storage capacity - up to 45TB for I7i and up to 120TB for I7ie instances, the highest available in the cloud. Both instance families feature always-on AES-256 encryption for enhanced security, and support Torn Write Prevention (TWP), enabling up to 30% better database performance.

I7i instances optimize transaction processing speeds for highly transactional workloads, enabling higher throughput and helping meet strict application service level agreements (SLAs). This increased performance allows you to deliver faster response times to end-users while processing more transactions per second.

With I7ie instances, you can reduce your cost per GB of SSD instance storage by up to 45% compared to I7i instances, allowing you to better right-size your instance selection based on your specific storage requirements.

I7i and I7ie instances are built on the AWS Nitro System, a combination of dedicated hardware and lightweight hypervisor which delivers practically all of the compute and memory resources of the host hardware to your instances for better overall performance and security.

Features

I7i and I7ie instances are powered by 5th generation Intel Xeon Scalable processors (Emerald Rapids) with an all-core turbo frequency of 3.2 GHz (max core turbo frequency of 4.0 GHz), and include support for always-on memory encryption using Intel Total Memory Encryption (TME).

I7i and I7ie instances use DDR5 memory, delivering enhanced bandwidth compared to previous generations. These instances support up to 100 Gbps of networking bandwidth and 60 Gbps bandwidth to HAQM Elastic Block Store (EBS). You can attach up to 128 EBS volumes to an instance. Additionally, the 48xlarge and metal-48xl sizes support Elastic Fabric Adapter (EFA) for high-performance computing applications.

5th Generation Intel Xeon Scalable processors feature four built-in accelerators: Advanced Matrix Extensions (AMX) for accelerating CPU-based machine learning through optimized matrix multiplication, Data Streaming Accelerator (DSA), In-Memory Analytics Accelerator (IAA), and QuickAssist Technology (QAT). DSA, IAA, and QAT, exclusively available on I7i and I7ie bare metal instances, optimize performance by efficiently offloading database operations, encryption, compression, and queue management tasks from the CPU.

Product details

I7i instances deliver the best compute and storage performance among x86-based HAQM EC2 storage instances. I7ie instances deliver the highest local NVMe storage density in the cloud.

  • I7i
  • Instance Size vCPU Memory (GiB) Instance Storage (GB) Network Bandwidth (Gbps) EBS Bandwidth (Gbps)
    i7i.large 2 16 1 x 468 GB = 468 GB Up to 10 Up to 10
    i7i.xlarge 4 32 1 x 937.5 = 937.5 GB Up to 10 Up to 10
    i7i.2xlarge 8 64 1 x 1875 = 1,875 GB Up to 12 Up to 10
    i7i.4xlarge 16 128 1 x 3,750 = 3,750 GB Up to 25 Up to 10
    i7i.8xlarge 32 256 2 x 3,750 = 7,500 GB Up to 25 10
    i7i.12xlarge 48 384 3 x 3,750 = 11,250 GB Up to 28.125 15
    i7i.16xlarge 64 512 4 x 3,750 = 15,000 GB Up to 37.5 20
    i7i.24xlarge 96 768 6 x 3,750 = 22,500 GB Up to 56.25 30
    i7i.48xlarge 192 1536 12 x 3,750 = 45,000 GB Up to 100 60
    i7i.metal-24xl 96 768 6 x 3,750 = 22,500 GB Up to 56.25 30
    i7i.metal-48xl 192 1,536 12 x 3,750 = 45,000 GB Up to 100 60
  • I7ie
  • Instance Size vCPU Memory (GiB) Instance Storage (GB) Network Bandwidth (Gbps) EBS Bandwidth (Gbps)
    i7ie.large 2 16 1 x 1,250 GB = 1,250 GB Up to 25 Up to 10
    i7ie.xlarge 4 32 1 x 2,500 GB = 2,500 GB Up to 25 Up to 10
    i7ie.2xlarge 8 64 2 x 2,500 GB = 5,000 GB Up to 25 Up to 10
    i7ie.3xlarge 12 96 1 x 7,500 GB = 7,500 GB Up to 25 Up to 10
    i7ie.6xlarge 24 192 2 x 7,500 GB = 15,000 GB Up to 25 Up to 10
    i7ie.12xlarge 48 384 4 x 7,500 GB = 30,000 GB Up to 50 15
    i7ie.18xlarge 72 576 6 x 7,500 GB = 45,000 GB Up to 75 22.5
    i7ie.24xlarge 96 768 8 x 7,500 GB = 60,000 GB Up to 100 30
    i7ie.48xlarge 192 1,536 16 x 7,500 GB = 120,000 GB 100 60
    i7ie.metal-24xl 96 768 8 x 7,500 GB = 60,000 GB Up to 100 30
    i7ie.metal-48xl 192 1,536 16 x 7,500 GB = 120,000 GB 100 60

Customer stories

  • Brave

    Brave Software, developer of a privacy-focused browser and search engine, has achieved significant performance improvements by migrating their search engine workloads to HAQM EC2 I7ie instances. This strategic infrastructure change resulted in a 3X improvement in search performance and a 30% reduction in latency, enabling Brave to process more user queries while maintaining their commitment to user privacy. By leveraging AWS's compute infrastructure through I7ie instances, Brave has positioned itself to scale its search engine operations efficiently while continuing to deliver fast, private search results for its growing user base.

    Subu Sathyanarayana, VP of Engineering, Brave Search

  • Elastic

    Elastic is committed to enabling customers to drive innovation and cost-efficiency with our Search AI powered observability, security, and search solutions on AWS. Our testing of the new i7i instances demonstrated up to 50% improved indexing throughput compared to the previous i4i generation. We are looking forward to the introduction of i7i instances, which promises to provide improved performance for Elasticsearch users deploying on AWS.

    Yuvraj Gupta, Principal Product Manager, Elastic

  • Nutanix

    Nutanix is excited to expand our collaboration with AWS by bringing Nutanix Cloud Clusters (NC2) to the I7i metal instances. With advanced performance, increased memory bandwidth, and flexible storage, these instances align perfectly with our vision of delivering high-performance hybrid cloud solutions at a lower TCO. With NC2 on AWS, customers can now leverage the power of Nutanix’s industry-leading hyperconverged infrastructure on the latest AWS hardware, ensuring greater scalability, resilience, and efficiency for their most demanding workloads. Additionally, the built-in Intel Advanced Matrix Extensions (AMX) and Nutanix AI with CPU based Machine Learning further optimizes performance, enhancing decision-making for complex applications. We look forward to continuing our partnership with AWS to enable enterprises to accelerate cloud adoption with simplicity and flexibility.

    Saveen Pakala, VP Product Management Nutanix