Home > Workload Solutions > Oracle > White Papers > MySQL InnoDB Cluster on Dell EMC PowerStore T > Storage design
Storage Pools enable the business to intelligently place workloads based on the optimal storage type. PowerStore Storage Pools can be used for data segregation, tiering, and performance management. PowerStore also features the flexibility to non-disruptively move storage pools from one storage type to another.
The PowerStore storage platform was designed to include native integration with Non-Volatile Memory Express (NVMe) drives and Intel Optane Storage Class Memory. For example, an Oracle database supporting enterprise applications will achieve optimal storage performance when using Intel Optane or NVMe storage.
The complete list of supported drives also includes SAS SSDs. The following table compares the full set of drive type characteristics that are available for PowerStore:
Drive type |
Benefits |
Intel Optane Storage Class Memory |
Exceptionally low read latency Maintains low read latency under heavy write workloads |
NVMe SSD |
Low read latency Higher operations per second |
SAS SSD |
Balanced performance and capacity |
Serial-attached SCSI (SAS) Solid State Disk (SSD) drives use the legacy SCSI command set protocol. This command set worked well with traditional hard drives; however, as technology advanced, the protocol became inefficient for faster I/O processing and was surpassed by newer protocols with less overhead.
Nevertheless, SAS SSD drives can deliver cost-effective performance for many applications at a lower cost per GB than NVMe SSD and Optane Storage class memory (SCM) drives. Consider SAS SSD drives for database environments where performance is not a priority and for those that would benefit from greater total available capacity in a limited data center space.
NVMe SSD drives can deliver higher data transfer rates than either Serial AT Attachment (SATA) or SAS interfaces because they offer more available channels (lanes) through which data can flow. NVMe uses the PCIe bus and an optimized protocol that reduces overhead, further boosting performance over older technologies. Optimizations in the NVMe protocol include deep I/O queues and support for 64K commands per queue.
For heavy workloads, SSDs based on NVMe are more responsive and less prone to performance degradation due to large numbers of I/O requests. These factors are often an important benefit for database workloads where both high Input Output Operations per Second (IOPS) requirements and increased bandwidth demand scenarios need consistent low latency response.
Organizations are increasingly choosing NVMe SSD as the most cost-effective storage for high-performance database platforms and similar business-critical applications. Consider the benefits of using NVMe SSDs for production databases where enterprise performance is critical.
Intel Optane SCM is a new class of memory that brings storage performance closer to DRAM memory speeds. DRAM is still many times faster, but the benefit of SCM memory is its persistence. SCM memory can be used as storage to provide a faster tier of performance. With SCM memory, read latencies are very low and are less affected by writes to the storage. SCM memory is best positioned for production and real-time analytics that require the best possible performance.
PowerStore implements a Dynamic Resiliency Engine (DRE) that is based on a proprietary algorithm. The software creates appropriate redundancy by automatically consuming all the drives within an appliance intelligently without any user intervention. Data, parity, and spare capacity are distributed across all the drives within an appliance and data is protected using an advanced methodology.
Data efficiency is enabled for a PowerStore array through its built-in data deduplication and data compression features. The PowerStore array with modern Intel Xeon processors is designed for low-overhead, always-on data deduplication and compression. These data reduction technologies have no perceptible impact on storage performance. Both these data reduction features are always-on for the array and cannot be disabled by storage administrators.
Data deduplication is a technology that identifies redundancies in the data that is stored to the system. Because PowerStore uses an efficient system of storage location “pointers,” a single copy of the redundant data can be referenced by multiple volumes, which saves space. The process of detecting and managing data deduplication occurs before the data is written to the storage systems. Scanning before storing reduces the number of write I/O operations, compared to deduplication methods that periodically scan the data after it is written to disk. Benefits of data deduplication include:
Data compression is the process of reducing the size of data stored to the system by implementing a more compact representation without losing information content. Customers gain additional value from PowerStore when the system can efficiently store more data on fewer blocks. Like deduplication, compression also reduces the number of write operations to the storage drives, which increases both reliability and performance.
Storage administrators can easily track the physical capacity utilization (used and free space) and data savings statistics in the PowerStore Manager dashboard under the Capacity tab. Data savings metrics are global for the array, which means that the overall efficiency, space savings, and space savings ratio describe everything on the storage system. This is primarily the result of how global deduplication works.
The following table shows the PowerStore T configuration that this MySQL solution uses. We implemented high capacity NVMe drives to provide a total capacity of 28 TB of storage capacity. Because this Power T configuration was designed to balance performance and capacity, the focus of this MySQL solution is to validate the infrastructure.
Feature |
PowerStore T capability |
PowerStore OS |
Operating system version 1.0.1.0.5.002 |
Processor |
4 x Intel Xeon Silver 4108 CPU @ 1.80GHz, 16C/32T per node |
Memory |
192 GB per node and 2 x 8 GB NVMe NVRM for caching |
Drives |
22 NVMe SSD drives each 1.92 TB |
Total capacity |
28.3 TB |