What is NVDIMM and how does it differ from memory channel flash products? How does NVDIMM work -- what allows it...
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
to retain memory without power?
NVDIMM is a form of non-volatile DRAM also called "NVRAM" or "NVDRAM." In-memory storage is non-volatile flash storage. NVDIMMs look, feel and act like DDR3 DRAM plugging into standard DIMM slots. The performance is the same as DRAM.
In-memory flash storage, on the other hand, connects as a DDR3 DIMM, but it is storage, not memory. It is designed to offer lower latency than PCIe flash storage because it does not have to traverse the PCIe controller or contend with other cards for bandwidth on the PCIe channel. In-memory flash storage is currently only available from SanDisk as ULLtraDIMM. SanDisk has an exclusive agreement with Diablo Technologies that pairs their DDR3 DIMM-based flash storage to SanDisk's Guardian software.
NVDIMMs are DRAM that is made non-volatile by utilizing a battery backup or super-capacitor, or "supercap." The battery backup or supercap provides the NVDIMM time, in the event of a power failure, for the NVRAM to transfer their data to flash or HDDs. Data is preserved even though the power has failed.
NVDIMM capacities are relatively small at 4 GB, 8 GB and 16 GB. Those small capacities limit scalability. Additionally, small capacities and non-volatility makes NVDIMMs more expensive than standard SDRAM per GB, and a lot more expensive than in-memory flash storage that comes in 200 GB and 400 GB capacities. However, NVRAM latency is equal to DRAM. And that latency is approximately 90% less than in-memory flash storage.
Comparing DIMM-connected flash vs. PCIe flash benefits
Solid-state memory channel storage takes on latency issues
Smart Storage goes DIMM with memory channel storage
Comparing performance and cost of DRAM vs. NAND
Related Q&A from Marc Staimer
Latency in object stores that stems from a large amount of metadata means the technology is better suited to non-transactional data.continue reading
Eventual consistency in object stores can be an issue because object storage is spread over many nodes and up-to-date data may not always be ...continue reading
HDD failure can put bytes of data at risk. Is multi-copy mirroring or erasure coding the more efficient data protection approach?continue reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.