An Industry First
On May 21st, Micron began shipping and selling the industry’s first QLC SSD, delivering the next evolution in storage that delivers fast capacity for less. Micron’s QLC technology delivers the features, capacity, performance and value to challenge HDDs in environments where performance is critical and the cost of SSDs has historically been too high. With QLC SSDs, you can quickly access and analyze the read-intensive, performance-sensitive workloads that fuel your business — like real-time analytics, machine learning, artificial intelligence, Big Data, content delivery, user authentication, and more.
Top Three Benefits of QLC SSDs
1. Lower TCO: While performance focused, read-centric workloads rely on massive arrays of HDD to deliver the results that users demand, QLC storage does it with fewer drives — lowering your performance-focused total cost of ownership (TCO).*
2. More Capacity: QLC stores more bits per cell than its predecessors, enabling immense gains at the system, rack and data center levels that are more cost-effective.
3. Smaller Footprint: QLC technology’s ability to increase per-server density reduces rack space by up to 7.7X compared to 2TB HDDs, saving precious and expensive data center real estate.
*All comparisons based on a 3.5-inch 8TB 7200RPM HDD. TCO statement based on common read-centric workload of 128K 95% read, estimations for similar performance (total capacity is different); 3X lower power comparisons based on public power consumption specifications for HDDs and Micron 5210 enterprise QLC SSD; 41% less rack space based on 42U rack (fully populated), 50PB data set, 2.5-inch 7.68TB Micron 5210 enterprise QLC SSDs, 8TB 3.5-inch enterprise SATA HDDs and high-density 2U servers capable of supporting 32X 2.5-inch or 18X 3.5-inch form factor devices per chassis, RAID 0
Ideal Applications for QLC Enterprise SSDs
Give these applications a performance uplift that’s more cost-effective using high-capacity, lightning-quick QLC-based SSDs.
- Real-Time Analytics and Big Data: Deliver value to Big Data applications like Hadoop Distributed File Systems.
- Business Intelligence and Decision Support Systems: Quickly mine massive data sets using faster, deeper queries for real-time business insights and decision-making.
- Active Archives and Large Block Storage: Transform scale-out active archives into a strategic asset and deliver massive large-block data streams with ease.
- Read-Intensive Artificial Intelligence: Get the speed AI algorithms depend on to quickly identify patterns in sprawling data sets.
- Machine and Deep Learning: Crunch immense data sets in seconds versus minutes — because machines can only learn as fast as they can read and analyze data.
- Content Delivery, Video on Demand, Media Streaming: Deliver more assets to more users more consistently with support for massive, parallel requests and streams.
- NoSQL Databases: Breathe life into data-driven workloads like content classification and tagging as well as user profile acceleration.
- User Authentication: Perform quick authentication with quick storage.
|Workload||Micron 9300 (TLC)||Micron 5210 (QLC)|
|AI & Machine Learning||In-node direct-attached storage (caching training data)||Centralized storage for the 50TB+ data lakes that feed model training & development|
|NoSQL databases (Cassandra, MongoDB)||Accelerate write-intensive & mixed-use workloads||HDD displacement for read-intensive workloads|
|Big Data & real-time analytics (Hadoop)||YARN caching||HDFS HDD displacement for faster time-to-insight|
|Block & object stores (Ceph)||Small random block storage||Object stores & media streaming|
|vSAN||Caching tier||Capacity tier|
|SQL databases||OLTP||BI/DSS (business intelligence)|
|High-frequency trading||Comparative analytics, transient dataset capture, complexity reduction (systems, drives, support)||HDD displacement for back-testing, regulatory/compliance reporting, and storing datasets that must be retrieved quickly|
How much is your time worth?
Following table is intended to give you an indication of how time-efficient the replacement of conventional HDDs by QLC SSDs is.
|ML Workflow||7.2K HDD (8TB)||5210 SSD (7.68TB)||Savings|
|# of Images||Dataset Size||Completion Time||Completion Time||Time Saved||Efficiency Increase|
|1000||23GB||10 minutes||1 minute||9 minutes||10x|
|10.000||230GB||90 minutes||11 minutes||1 hour 19 minutes||9x|
|50.000||1.150GB||454 minutes||54 minutes||6 hours 40 minutes||8x|
|100.000||2.300GB||910 minutes||113 minutes||13 hours 17 minutes||8x|
- The more ML you do, the more you compound your time savings
- The bigger your dataset, the more time you can save
- Storage is often a bottleneck in ML workloads, as the size of the dataset often dwarfs the installed memory capacity
- The long-term costs of underperforming GPU/CPUs often far outweigh the cost of QLC SSDs
- The Micron 5210 is 3x more energy efficient so you save on power and cooling, too>