Storage without limit. Accelerate your business with Phalanx.

Business and ideas don't wait. Neither should you. Moving at the speed of business demands the best from your IT infrastructure. Phalanx offers the agility, resiliency, and performance to keep your organization growing.


Up to

16x

more IOPS1

With an innovative row & column architecture that evenly distributes workloads across devices and capacity.


Up to

19%

faster throughput2

Using customized device-aware data layout that reduces contention and maximizes concurrency.


Up to

58%

more transactions3

Powered by an AI-enhanced stack that intelligently prioritizes each IO and optimizes every data placement.


Up to

10x

better performance consistency3

By incorporating a specialized metadata path that enhances the scalability, efficiency and robustness of the system.


Up to

4.8x

lower latency4

Through tailored device-friendly IO that reduces device workload, increases endurance, and cuts long tail latency.

All the power you need. In a system that is easy and simple to use.



The power of next-gen storage, now at your fingertips. Seamless support for new devices such as SMR, HAMR, and multi-actuator. A versatile software-defined model that fits every workflow and environment. Phalanx empowers you for what's next.

Uncompromising reliability and performance with effortless simplicity. A powerful combination.

Phalanx delivers the best performance of any storage system. Start focusing on what you are creating, not what you are creating with.



Making IO fast and resilient.

Phalanx intelligently prioritizes incoming requests and data placement — enabling concurrent streams of high throughput data-intensive and low latency interactive I/O. A dedicated metadata store further increases the scalability, efficiency, and resiliency of your data.


Every device fits just right.


Phalanx tailors access patterns to match device capability, empowering peak performance. Built on a game-changing row & column architecture and powered by an AI-enhanced stack — Phalanx enables you to store and access data in ways never before possible.

Up to

4.8x

lower latency4

Percentile latencies (usec)

Kalista Phalanx
with
Ultrastar HC620

Legacy stack
with
Ultrastar HC530

99%

16,924

28,468

99.95%

26,211

97,371

99.99%

41,736

202,227

Increasing density and lowering TCO.

While HDD capacity has increased over time, the rate of IO processing has not kept pace. This asymmetry leads to cost and scalability issues. Phalanx is designed from the ground up to address these challenges and to return the power to the user. This means more time and resources for what matters most - doing your best work.


Up to

16x

more IOPS1

An industry first. Host Managed SMR on Ceph and Hadoop. Say hello to consistency, predictability, and a lower TCO.

The most advanced storage platform, packaged with a sleek, simple turnkey design that is easy to deploy and operate. There are no application nor kernel changes. We've minimized the friction and disruption to what's next.



Let the yellow elephant out. Hadoop does not natively support next generation storage technologies such as Host Managed SMR. Resulting in higher TCO and inconsistent performances at scale. With Phalanx, Hadoop can transparently utilize Host Managed SMR devices. Users no longer have to compromise on performance for increased storage density and lower TCO.

Up to

19%

faster throughput2

Release the Kraken. Ceph is not designed to minimize IO contentions and hot writes. These lead to scalability issues and long tail latencies. Phalanx enables Ceph to scale performance with capacity and to limit performance variations at scale. It’s now easier than ever to expand your storage.


Up to

58%

higher IOPS3


Up to

10x

better performance consistency3

Performance and simplicity. All together now.


Learn how Phalanx is enabling optimal performance and TCO at scale.


Phalanx architecture

Flexible I/O tester (fio) benchmarks

Hadoop TestDFSIO benchmarks

Ceph OSD and Rados benchmarks

Kalista IO and Western Digital joint solution brief

Phalanx is engineered to grow with you. This means predictable and consistent performance at any scale. So that you can do more with less. From here anything is possible.

Interested to know more? We invite you to connect with us to discuss your business requirements and see if Phalanx is a good fit for your organization.
Contact us at info@kalista.io

  1. Testing conducted by Kalista IO in August 2019 using preproduction Olympus (Phalanx) software with Linux kernel 4.18.0-25-generic, and Intel® Core™ i7-4771 CPU 3.50GHz with 16GiB DDR3 Synchronous 2400 MHz RAM, and Western Digital Ultrastar DC HC620 host managed SMR and Ultrastar DC HC530 CMR drives connected through SATA 3.2, 6.0 Gb/s interface. Tested with Flexible I/O tester (fio) version 3.14-11-g308a. Random write bench ran for 1800 seconds with 4KB block and 200GB file size, 64 concurrent threads each with queue depth of 1. Executed 3 times to capture average and standard deviation IOPS values.
  2. Testing conducted by Kalista IO in August 2019 using preproduction Olympus (Phalanx) software with Linux kernel 5.0.0-25-generic, and Intel® Core™ i7-4771 CPU 3.50GHz with 16GiB DDR3 Synchronous 2400 MHz RAM, and Western Digital Ultrastar DC HC620 host managed SMR and Ultrastar DC HC530 CMR drives connected through SATA 3.2, 6.0 Gb/s interface. Tested with Apache Hadoop version 3.2.0 in single node pseudodistributed mode with single block replica, and TestDFSIO version 1.8 on OpenJDK version 1.8.0_222. TestDFSIO read benchmark ran with 32 files, 16GB each for a 512GB dataset. Executed 3 times to capture average and standard deviation throughput values.
  3. Testing conducted by Kalista IO in August 2019 using preproduction Olympus (Phalanx) software with Linux kernel 5.0.0-25-generic, and Intel® Core™ i7-4771 CPU 3.50GHz with 16GiB DDR3 Synchronous 2400 MHz RAM, and Western Digital Ultrastar DC HC620 host managed SMR and Ultrastar DC HC530 CMR drives connected through SATA 3.2, 6.0 Gb/s interface. Tested with Ceph version 13.2.6 Mimic in single node mode with single object replica. Rados write bench ran with 4MB object and block (op) size with 16 concurrent operations for 1800 seconds to capture average and standard deviation IOPS values.
  4. At 99.99th percentile latency. Testing conducted by Kalista IO in July 2020 using preproduction Olympus (Phalanx) software and XFS file system with Linux kernel 5.4.0-42-generic, and Intel® Core™ i7-4771 CPU 3.50GHz with 16GiB DDR3 Synchronous 2400 MHz RAM, and Western Digital Ultrastar DC HC620 host managed SMR and Ultrastar DC HC530 CMR drives connected through SATA 3.2, 6.0 Gb/s interface. Write bench created a single 1GB file and executed 600,000 write commands each overwriting the first 64KB region of the file capture latency values.

"They can conquer who believe they can." — Virgil