Among all the layers from the Epic electronic medical record (EMR) solution, the InterSystems Caché® database subsystem is more likely to affect the overall user experience – so it’s critical for the design to follow the guidelines for Caché. The table below is an example of those guidelines.1
- For randomly placed reads to database files measured at the system call level:
o Average read latencies must be 2 ms or less
o 99% of read latencies must be below 60 ms
o 99.9% of read latencies must be below 200 ms
o 99.99% of read latencies must be below 600 ms
- For randomly placed writes to database files measured at the system call level:
o Average write latencies must be 1.0 ms or less
As you can see, there are guidelines on both read and write latency. Deployments must meet both values (average and maximum response time). Databases are a tough customer for storage as they have multiple types of I/O profiles with various needs and requirements within the solution.
One of the most difficult metrics to achieve, which is critical for the customer experience, is to keep read response times low while performing write operations. If you have been around databases, you know that checkpoint, logs, and journals are occurring consistently. With the Caché database, every 80 seconds a single threaded 256KB block size sequential write happens. This type of transaction is exactly the use case for which the Intel® Optane™ DC SSD P4800X was created--to sustain high frequency flushes while maintaining predictable response time.
What does it mean to YOU?
A solution with an Intel® Optane™ DC SSD in the caching tier will deliver the low latency needed for a database, such as Caché, for its most critical operations. And your Epic EMR solution will have an excellent user experience across the layers.
The graph above compares an Intel® Optane™ DC SSD with an Intel® 3D NAND SSD2. The 3D NAND drive is commonly used as a capacity drive. In this write pressure test, the data shows how each drive handles read latency as we increase write pressure. You can see that the Intel Optane DC SSD performs at 63X lower maximum latency against the Intel 3D NAND SSD. This reduced latency is the key to a predictable database response. As a result, putting Intel® Optane™ DC SSDs in the cache layer gives you the confidence to deploy multiple workloads on your HCI solution, and meet the challenging database requirements. Read more about the performance of an Intel® Optane™ technology cache for databases in this paper from ESG.
So, when you’re deploying an EMR database and latency really matters, you’ll need a solution that can perform under the stringent requirements. The Cisco HyperFlex All NVMe configuration with Intel® Optane™ technology was recently tested and exceeds Epic’s operational database performance requirements. Read Cisco’s blog about the HyperFlex end-to-end Epic solution.
For more about Intel and Intel® Optane™ technology, please see links below and your Intel representative.
• Intel.com: Intel® Optane™ Technology
• Product Brief: Intel® Optane™ SSD DC P4800X/P4801X
• Cisco Blog: Cisco Hyperflex Now Supports End-End Epic EMR
• Citrix Blog: See How Better Becomes Possible with Citrix and Cisco HyperFlex
• Enterprise Strategy Group Testing of HyperFlex: Mission-Critical Hyperconverged Workload Performance Testing Cisco HyperFlex All NVMe with Intel Optane SSD
1 Please note that these figures are provided as guidelines, and that any given application may have higher or lower tolerances and thresholds for ideal performance. These figures and I/O profiles are to be used as a starting point for your discussions with your storage vendor.
2 Source – Intel-tested: Average read latency measured at queue depth 1 during 4k random write workload. Measured using FIO 3.1. Common Configuration - Intel 2U Server System, OS CentOS 7.5, kernel 4.17.6-1.el7.x86_64, CPU 2 x Intel® Xeon® 6154 Gold @ 3.0GHz (18 cores), RAM 256GB DDR4 @ 2666MHz. Configuration – Intel® Optane™ SSD DC P4800X 375GB and Intel® SSD DC P4600 1.6TB. Latency – Average read latency measured at QD1 during 4K Random Write operations using FIO 3.1. Intel Microcode: 0x2000043; System BIOS: 00.01.0013; ME Firmware: 04.00.04.294; BMC Firmware: 1.43.91f76955; FRUSDR: 1.43. SSDs tested were commercially available at time of test. Performance results are based on testing as of July 24, 2018 and may not reflect all publicly available security updates. See configuration disclosure for details. No product can be absolutely secure.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit www.intel.com/benchmarks.