Good news for healthcare: HPE GreenLake for Block Storage MP validated with Epic Systems

[StorageExperts]

Learn about the validation of HPE GreenLake for Block Storage built on HPE Alletra Storage MP with Epic Systems.

– By Nick Jennings, Lead Engineer, Storage Solutions Engineering, HPE

Healthcare services typically rely on their electronic health records (EHR) system for access to their patient-centric data that needs to be available 24x7x365. Epic Systems develops and sells EHR software solutions that cover over 300 million patients worldwide. Epic EHR software solutions are used in most top hospitals and medical schools.

Healthcare providers are also demanding greater performance across the growing volume of patient data while reducing total cost of ownership (TCO) to maximize the return on their investment (ROI). The solution? HPE GreenLake for Block Storage powered by HPE Alletra Storage MP with Epic Systems software.

HPE GreenLake for Block Storage built on HPE Alletra Storage MP leverages a disaggregated, scale-out storage architecture for mission-critical workloads that enables scaling capacity and performance independently to increase efficiency and reduce costs. This provides flexibility to fine-tune, adapt, and elastically scale storage environments in line with workload requirements and evolving healthcare needs.

Validation environment details

The system validated was an entry-point switchless HPE GreenLake for Block Storage MP with two nodes, dual CPUs with 32 cores, and 24 2 TB NVMe drives, running HPE GreenLake for Block Storage OS Release 4. Any configuration of HPE GreenLake for Block Storage MP with more nodes, drives, and/or switches would provide similar or better performance and are also considered validated.

Twelve volumes from the storage were exported to the VMware ESX® server and spread evenly across four virtual SCSI controllers using physical raw device mapping (RDM). The storage was then configured on the host by creating volume groups, logical volumes, and XFS file systems.

The test host was a Red Hat® Linux® Enterprise 8 VM running on ESX 8, configured with 16 CPU cores and 64 GB of RAM. It was connected to the storage through a 32 Gb Fibre Channel switch.

Test description and results

An Epic Systems load simulator measures and reports the storage IOPS capability, which is determined by a 45-second Max Write Time threshold. The Max Write Time is the longest any write took during the cycle. For each test, twenty cycles were run with increasing levels of IOPS. 

Each of these three tests was run twice, one with volumes with deduplication and compression (DECO) and again with thin provisioned virtual volumes (TPVV). 

• Base volumes test – This measured performance without any snapshot activity.
• Scheduled snapshots test – This measured performance with snapshots scheduled through the protection policy scheduler in Data Services Cloud Console. Snapshots were taken every 30 minutes with a retention of four hours when the snapshots would be expired. Once per hour, expired snapshots were deleted from disk.
• Scripted snapshots test – This measured the performance with snapshots taken through a script that utilized the updatevv CLI command. Snapshot volumes (vcopy volumes) were attached to the base volumes, and the script utilized the updatevv command every 30 minutes to create an updated snapshot on the attached volumes.

The June 2024 release of the Epic Systems Storage Products and Technology Status guide reflects the validation of HPE GreenLake for Block Storage built on HPE Alletra Storage MP Release 4 and later.

To learn more, check out the technical brief with more information about the validation of HPE GreenLake for Block Storage built on HPE Alletra Storage MP with Epic Systems.

---

Meet HPE Storage Experts blogger Nick Jennings

Nick is the lead engineer on the worldwide Storage Solutions Engineering team at HPE. His focus is on storage technology in the healthcare industry.

Storage Experts
Hewlett Packard Enterprise

twitter.com/HPE_Storage
linkedin.com/showcase/hpestorage/
hpe.com/storage