Intel X25-M 80GB SSD, Intel Ups The Ante

Article Index:   

Anatomy Of The X25-M 80GB SATA Solid State Disk



Intel's new line of SSDs looks pretty much like any other Solid State Drive on the exterior, but it's the interior components that make up Intel's "special sauce".

 

 

The outside label of the Intel X25-M 80GB SSD shows a model number of SSDSA2SH080G1GN and it's described as a 2.5", 3GB/s SATA SSD 80GB, 5V, 1A.  As you can also see, we have an engineering sample here that is Pb-free (lead free, RoHS compliant) and made in China.

Under the hood is where all the magic happens, and as you can see save for one 128Mb Samsung DRAM chip, this SSD is all Intel inside.  The Samsung DRAM is a rather shallow cache buffer at 128 megabit versus megabytes.  This memory is temporary scratch-pad storage for Intel's flash memory controller ASIC.  This chip provides access to 10X NAND Flash channels and offers features like Native Command Queuing and up to 32 concurrent read/write operations.  In addition, this chip also offers a standard wear-leveling algorithm so that the erase/re-write cycle endurance of the drive is evenly spread across all memory locations. 

Intel's Adaptive Memory Controller Technology
Finally, and most importantly, Intel's flash memory controller offers dynamic workload adaptation such that the drive will actually adjust on the fly to specific workload conditions.  For example, install this drive into a laptop and over time it will adjust and optimize to the OS and standard application workload and access pattern requirements for the most active file sizes and read/write operations you run.  If you then re-purpose the drive into another type of installation or dramatically change your traditional storage workload, the drive will have to re-adapt to your new usage model and in the interim, performance will be degraded temporarily (we actually witnessed this in a benchmark run).  We should note that a user's workload usage model would have to change fairly significantly for it to affect short-term performance.  Regardless, this is a performance optimization that you will see pays dividends in our upcoming benchmark runs.
 

Image gallery

Related content

Comments

blog comments powered by Disqus