StorPool on AWS: StorPool Storage Enables Performance Intensive Workloads in Amazon AWS

Katerina Temelkova Blog Leave a Comment

In the past decade, hyperscale public clouds have made it possible to build large-scale cloud-native applications by taking advantage of immense horizontal scalability. However, a substantial set of enterprise applications have historically relied heavily on vertical scaling (scale up). This makes it hard to lift and shift these applications to any hyperscale cloud. In particular, cloud-based instances only support a modicum of storage performance, and often traditional applications demand much more storage performance per instance than any cloud vendor can provide.

StorPool delivers more than 5 times the performance of Amazon EBS while reducing latency by 50%

With the StorPool on AWS technology solution, companies can benefit from world-class single-instance storage performance, matching or exceeding the storage performance of high-end on-premises block storage. With the presented approach, StorPool enables the deployments of heavy enterprise applications that are not economically achievable with other technologies. 

The StorPool solution was tested and benchmarked on AWS in cooperation with Randy Seamans, a Global Storage Solutions Specialist at Amazon Web Services. The solution makes a new block-level software defined storage offering available on AWS with industry-leading performance and low latency, while providing enterprise- grade features, like instant no-performance-impact snapshots and thin clones.

The StorPool on AWS solution is a pre-architected, easy-to-deploy, stable, and full-featured solution. It passed all load and functional testing and we are now looking for partners to deploy the StorPool on AWS solution for applications and use-cases which require exceptional storage performance of more than 1 million balanced random read/write IOPS to a single EC2 r5n instance. 

To start your StorPool on AWS deployment, write us an email: [email protected].

StorPool Storage now runs on Amazon Web Services

StorPool Storage is the best software-defined primary storage platform for large-scale cloud infrastructure, due to cost per IOPS, storage efficiency, advanced features, and optimal performance even when engaging snapshots and other features that decrease performance in other storage architectures. Traditionally, customers deploy StorPool on standard servers in their datacenter space. StorPool storage systems are ideal for storing and managing the data of workloads that demand extreme reliability and ultra-fast data access times – databases, web servers, virtual desktops, real-time analytics solutions, and other mission-critical software.

We are excited to announce that starting in August 2022 it’s possible to perform Proof of Concept deployments of StorPool Storage on i3en.metal instances in AWS. Users can get the same white-glove service provided to our on-premises customers and evaluate StorPool’s fit in their hybrid cloud strategy. Our team of seasoned storage experts will architect, deploy, tune, monitor, and maintain each StorPool storage system, so that users can focus on running their business with the peace of mind that their application’s foundation is running optimally. 

StorPool Storage delivers blazing-fast 1M+ balanced random read/write IOPS to a single EC2 r5n instance. StorPool frees you of per-instance storage limitations and can deliver this level of performance on any compatible instance type with sufficient network bandwidth. It achieves these numbers while utilizing less than 17% of client CPU resources for storage operations, leaving the remaining 83% for the user application(s) and database(s). 

To illustrate the “performance envelope” of a small StorPool on AWS system, this chart shows latency vs. IOPS of 4KB mixed read/write storage operations on an r5n client instance. The StorPool storage system, when running on 5x i3en instances delivers more than 1,200,000 IOPS at very low latency, compared to io2 Block Express which tops out at about 260,000 of the same type of IOs.

In terms of storage throughput, StorPool can deliver more than 10 GB/s to a single client instance. StorPool saturates 100 Gbps network connections with storage operations with block size of 8KB or larger.

And, most incredibly, this level of performance is achieved while keeping the complete set of benefits from StorPool’s advanced features. With StorPool, you get a linearly scalable primary storage system with end-to-end data integrity, instantly available snapshots & instant space-efficient clones with multi-region or availability zone snapshot transports, the ability to add storage nodes while the storage system is online (which adds capacity and performance), and non-disruptive software updates and upgrades.

Technical Details about the implementation:

AWS’s unique technologies assist and enhance the capabilities of the StorPool data storage platform:

  • i3en.metal instances provide 8 x NVMe SSDs direct-attached per instance (60 TB raw NVMe storage) and are the ideal building block for high-performance storage systems
  • 100Gbps Elastic Fabric Adapter (EFA) with libfabric (RDMA) is the low latency, high throughput network used in StorPool on AWS
  • Intel IOAT/QuickData Technology – the i3en.metal instance type supports Intel’s QuickData platform technology. StorPool on AWS uses QuickData for lower CPU usage, lower latency, and increased storage throughput.
  • Wide range of compatible CPU-optimized, memory-optimized and general-purpose instance types that can leverage StorPool volumes: m5n, m5zn, m6i, m6a, c5n, c6a, c6i, r5n, r6i, or any other EFA-enabled instances can act as storage clients (i.e., run customer workloads) 

Use Cases for StorPool Storage on Amazon AWS

The new solution is designed to achieve extremely low latency and high IOPS, delivered to single-node workloads such as large transactional databases, monolithic applications, and heavily loaded e-commerce websites. In particular, it is an excellent choice for traditional applications that previously might not have been possible on the public cloud. In addition, workloads that require extreme bandwidth block performance can leverage StorPool to deliver more than 10 GBytes/s of large block IO to a single client instance. Several times more throughput can be delivered from a StorPool storage system when serving multiple clients.

A perfect example is electronic health records (EHR) systems used by hospitals and large medical practices. These applications are typically run on a traditional architecture that requires healthcare providers to maintain two locations running on expensive high-end hardware. Large-scale EHR deployments require very high performance with low latency, which is a challenge even for on-premises deployments. With StorPool’s solution, it is easier to move at least one of those locations to AWS and decrease capital expenditures in favor of the more flexible OpEx model offered by the public cloud, potentially allowing healthcare organizations to eliminate a secondary datacenter entirely.

If you work at a healthcare provider, financial services, insurance company, SaaS vendor, or someone else taking care of monolithic applications that are hungry for performance, you can get started with an evaluation of StorPool on AWS right now. We will assist with setting up your application in AWS so you can get your storage performance hungry workloads running in the cloud, with the smallest possible footprint.

Supporting multiple use-cases from one storage system

Each StorPool Storage system can be used for multiple use cases using the same data thanks to the very high performance, scalable capacity, crash-consistent snapshots, and copy-on-write clones of the StorPool solution.

Multi-region architecture and DR

In addition, StorPool can be used in multi-region or multiple availability-zone deployments where data is periodically snapshotted and synchronized for secondary uses (Dev, Staging, Reporting/Analytics, Backups and DR). 

StorPool storage systems deployed on-premises can also send snapshots to StorPool in AWS and vice-versa, enabling and enhancing hybrid cloud deployments. 

Conclusion

Large applications with a traditional architecture, which were previously impossible to run in the public cloud can now run in Amazon AWS, by utilizing the StorPool on AWS solution.

For more information, questions or if you would like to test the StorPool on AWS solution – do contact us at [email protected]

About StorPool

StorPool Storage is a primary storage platform designed for large-scale cloud infrastructure. It is the easiest way to convert sets of standard servers into primary storage systems. The StorPool team has experience working with various clients – Managed Service Providers, Hosting Service Providers, Cloud Service Providers, enterprises and SaaS vendors. StorPool Storage comes as a software, plus a fully managed data storage service that transforms standard hardware into fast, highly available and scalable storage systems.

Under the hood, the primary storage platform is developed from the ground up and provides thin-provisioned volumes to the workloads and applications running in a cloud. The native multi-site, multi-cluster, and business continuity/disaster recovery capabilities supercharge hybrid- and multi-cloud efforts at scale. For more information about how StorPool powers the fastest public cloud in the world today, check out the 2022 public cloud performance measurement report.

Leave a Reply

Your email address will not be published.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.