Intel Officially Launches Optane Memory, Shows Performance

Subject: Storage | March 27, 2017 - 12:16 PM |
Tagged: XPoint, Optane Memory, Optane, M.2, Intel, cache, 3D XPoint

We are just about to hit two years since Intel and Micron jointly launched 3D XPoint, and there have certainly been a lot of stories about it since. Intel officially launched the P4800X last week, and this week they are officially launching Optane Memory. The base level information about Optane Memory is mostly unchanged, however, we do have a slide deck we are allowed to pick from to point out some of the things we can look forward to once the new tech starts hitting devices you can own.

Optane Memory-6.png

Alright, so this is Optane Memory in a nutshell. Put some XPoint memory on an M.2 form factor device, leverage Intel's SRT caching tech, and you get a 16GB or 32GB cache laid over your system's primary HDD.

Optane Memory-15.png

To help explain what good Optane can do for typical desktop workloads, first we need to dig into Queue Depths a bit. Above are some examples of the typical QD various desktop applications run at. This data is from direct IO trace captures of systems in actual use. Now that we've established that the majority of desktop workloads operate at very low Queue Depths (<= 4), lets see where Optane performance falls relative to other storage technologies:

Optane Memory-22.png

There's a bit to digest in this chart, but let me walk you through it. The ranges tapering off show the percentage of IOs falling at the various Queue Depths, while the green, red, and orange lines ramping up to higher IOPS (right axis) show relative SSD performance at those same Queue Depths. The key to Optane's performance benefit here is that it can ramp up to full performance at very low QD's, while the other NAND-based parts require significantly higher parallel requests to achieve full rated performance. This is what will ultimately lead to a much snappier responsiveness for, well, just about anything hitting the storage. Fun fact - there is actually a HDD on that chart. It's the yellow line that you might have mistook as the horizontal axis :).

Optane Memory-11.png

As you can see, we have a few integrators on board already. Official support requires a 270 series motherboard and Kaby Lake CPU, but it is possible that motherboard makers could backport the required NVMe v1.1 and Intel RST 15.5 requirements into older systems.

Optane Memory-7.png

For those curious, if caching is the only way power users will be able to go with Optane, that's not the case. Atop that pyramid there sits an 'Intel Optane SSD', which should basically be a consumer version of the P4800X. It is sure to be an incredibly fast SSD, but that performance will most definitely come at a price!

We should be testing Optane Memory shortly and will finally have some publishable results of this new tech as soon as we can!

Source: Intel

30 nanoseconds is way too slow, down with the latency gap!

Subject: General Tech | February 23, 2017 - 10:45 AM |
Tagged: hbll, cache, l3 cache, Last Level Cache

There is an insidious latency gap lurking in your computer between your DRAM and your CPUs L3 cache.  The size of the latency depends on your processor as not all L3 cache are created equally but regardless there are wasted CPU cycles which could be reclaimed.   Piecemakers Technology, the Industrial Technology Research Institute of Taiwan and Intel are on the case, with a project to design something to fit in that niche between the CPU and DRAM.  Their prototype Last Level Cache is a chip with 17ns latency which would improve the efficiency at which L3 cache could be filled to pass onto the next level in the CPU.  The Register likens it to the way Intel has fit XPoint between the speed of SSDs and DRAM.  It will be interesting to see how this finds its way onto the market.

dram_l3_cache_gap.jpg

"Jim Handy of Objective Analysis writes about this: "Furthermore, there's a much larger latency gap between the processor's internal Level 3 cache and the system DRAM than there is between any adjacent cache levels.""

Here is some more Tech News from around the web:

Tech Talk

Source: The Register

Intel Quietly Launches Official Optane Memory Site

Subject: Storage | February 15, 2017 - 08:58 PM |
Tagged: XPoint, ssd, Optane, memory, Intel, cache

We've been hearing a lot about Intel's upcoming Optane memory over the past two years, but the information had all been in the form of press announcements and leaked roadmap slides.

optane-memory-marquee-16x9.png.rendition.intel_.web_.1072.603.png

We now have an actual Optane landing page on the Intel site that discusses the first iteration of 'Intel Optane Memory', which appears to be the 8000p Series that we covered last October and saw as an option on some upcoming Lenovo laptops. The site does not cover the upcoming enterprise parts like the 375GB P4800X, but instead, focuses on the far smaller 16GB and 32GB 'System Accelerator' M.2 modules.

intel-optane-memory-8000p.jpg

Despite using only two lanes of PCIe 3.0, these modules turn in some impressive performance, but the capacities when using only one or two (16GB each) XPoint dies preclude an OS install. Instead, these will be used, presumably in combination with a newer form of Intel's Rapid Storage Technology driver, as a caching layer meant as an HDD accelerator:

While the random write performance and endurance of these parts blow any NAND-based SSD out of the water, the 2-lane bottleneck holds them back compared to high-end NVMe NAND SSDs, so we will likely see this first consumer iteration of Intel Optane Memory in OEM systems equipped with hard disks as their primary storage. A very quick 32GB caching layer should help speed things up considerably for the majority of typical buyers of these types of mobile and desktop systems, while still keeping the total cost below that for a decent capacity NAND SSD as primary storage. Hey, if you can't get every vendor to switch to pure SSD, at least you can speed up that spinning rust a bit, right?

Source: Intel

Full Steam ahead! Cache your games in this DIY cache server

Subject: General Tech | January 30, 2017 - 12:42 PM |
Tagged: steam, cache, Nginx, ubuntu

There are tricks to managing your Steam library if you are running low on space or simply setting up something new, from tricking Steam by copying files manually or the new feature which allows you to move games from within Steam.  One other possible way to manage your time and bandwidth is to build yourself a small little webserver which caches any Steam game you have downloaded locally, so you can reinstall them without using up your bandwidth.  Those familiar with Riverbed appliances and the like will already be familiar with this process but many gamers may not be.  Ars Technica walks you through the build and teaches a bit about caching and basic webservers along the way; check it out you are not already well versed in setting up something similar.

index.png

"But there’s an alternative to having to re-download all your Steam games from the Internet: you can set up a local Steam caching server, so that once you download something, you’ve got it on your LAN instead of having to reach for it across the net and incur usage fees."

Here is some more Tech News from around the web:

Tech Talk

Source: Ars Technica
Subject: Storage
Manufacturer: OCZ

Introduction, Specifications and Packaging

Introduction:

Since their acquisition by Toshiba in early 2014, OCZ has gradually transitioned their line of SSD products to include parts provided by their parent company. Existing products were switched over to Toshiba flash memory, and that transition went fairly smoothly, save the recent launch of their Vector 180 (which had a couple of issues noted in our review). After that release, we waited for the next release from OCZ, hoping for something fresh, and that appears to have just happened:

150707-180041.jpg

OCZ sent us a round of samples for their new OCZ Trion 100 SSD. This SSD was first teased at Computex 2015. This new model would not only use Toshiba sourced flash memory, it would also displace the OCZ / Indilinx Barefoot controller with Toshiba's own. Then named 'Alishan', this is now officially called the 'Toshiba Controller TC58'. As we found out during Computex, this controller employs Toshiba's proprietary Quadruple Swing-By Code (QSBC) error correction technology:

QSBC.png

Error correction tech gets very wordy, windy, and technical and does so very quickly, so I'll do my best to simplify things. Error correction is basically some information interleaved within the data stored on a given medium. Pretty much everything uses it in some form or another. Some Those 700MB CD-R's you used to burn could physically hold over 1GB of data, but all of that extra 'unavailable' space was error correction necessary to deal with the possible scratches and dust over time. Hard drives do the same sort of thing, with recent changes to how the data is interleaved. Early flash memory employed the same sort of simple error correction techniques initially, but advances in understanding of flash memory error modes have led to advances in flash-specific error correction techniques. More advanced algorithms require more advanced math that may not easily lend itself to hardware acceleration. Referencing the above graphic, BCH is simple to perform when needed, while LDPC is known to be more CPU (read SSD controller CPU) intensive. Toshiba's proprietary QSB tech claims to be 8x more capable of correcting errors, but what don't know what, if any, performance penalty exists on account of it.

We will revisit this topic a bit later in the review, but for now lets focus on the other things we know about the Trion 100. The easiest way to explain it is this is essentially Toshiba's answer to the Samsung EVO series of SSDs. This Toshiba flash is configured in a similar fashion, meaning the bulk of it operates in TLC mode, while a portion is segmented off and operates as a faster SLC-mode cache. Writes first go to the SLC area and are purged to TLC in the background during idle time. Continuous writes exceeding the SLC cache size will drop to the write speed of the TLC flash.

Read on for the full review!

Can Seagate's cached HDD compete with an SSD + HDD setup?

Subject: Storage | October 4, 2013 - 02:12 PM |
Tagged: hybrid hdd, cache, Seagate, 2TB

Benchmarking cached HDDs can be a difficult task as they are specifically designed to cache commonly used data which results in two very different speeds for data access, the 8GB SSD and the actual HDD.  The Tech Report recently met this challenge when benchmarking Seagate's first 3.5" desktop cached drive with 8GB of flash and 2TB of platter storage.  When contrasting it to some of the higher end HDDs available it became apparent that the more expensive WD Black 4TB was a faster drive but as it does cost more per gigabyte it might not be the best choice for every purpose.  Check out the review to see if this hybrid device is a better choice than buying both a small sized SSD and a large HDD for your own usage.

TR_drive.jpg

"Seagate's hybrid tech has finally been deployed in a desktop drive. The Desktop SSHD combines an 8GB flash cache with 2TB of mechanical storage. We take a closer look at how that combo holds up against standard hard drives and SSDs."

Here are some more Storage reviews from around the web:

Storage

HAMRs, Shingles and SSD cached HDDs; size versus speed

Subject: General Tech | August 8, 2013 - 01:58 PM |
Tagged: HAMR, SMR, cache, hdd, Seagate, western digital, hgst, helium

Enthusiasts are wholeheartedly adopting SSDs for their storage media of choice with HDDs relegated to long term storage of infrequently accessed storage.  For SMB and enterprise it is not such an easy choice as the expense to move to a purely SSD infrastructure is daunting and often not the most cost effective way to run their business.  That is why HDD makers continue to develop new technology for platter based storage such as HAMR and shingled magnetic media in an attempt to speed up platter drives as well as increasing the storage density.  Today at The Register you can read about a variety of technologies that will keep the platter alive, from Seagate's cached Enterprise Turbo SSHD, HGST's helium filled drives and the latest predictions on when HAMR and SMR drives could arrive on the market.

turbo_sshd_spc1c_iops.jpg

"At a briefing session for tech journos yesterday, Seagate dropped hints of new solid-state hybrid drives (SSHDs) - which combine a non-volatile NAND cache with spinning platters - and a general session about Shingled Magnetic Recording (SMR) and Heat-Assisted Magnetic Recording (HAMR)."

Here is some more Tech News from around the web:

Tech Talk

Source: The Register

New Samsung 840 EVO employs TLC and pseudo-SLC TurboWrite cache

Subject: Storage | July 18, 2013 - 01:12 AM |
Tagged: tlc, ssd, slc, sata, Samsung, cache, 840 evo

Samsung's release of the 840 EVO earlier today likely prompted some questions, such as what type of flash does it employ and how does it achieve such high write speeds. Here is the short answer, with many slides in-between, starting off with the main differences between the 840 and the 840 EVO:

DSC04627.JPG

So, slightly increased specs to help boost drive performance, and an important tidbit in that the new SSD does in fact keep TLC flash. Now a closer look at the increased write specs:

DSC04633.JPG

Ok, the speeds are much quicker, even though the flash is still TLC and even on a smaller process. How does it pull off this trick? Tech that Samsung calls TurboWrite.

DSC04637.JPG

A segment of the TLC flash is accessed by the controller as if it were SLC flash. This section of flash can be accessed (especially written) much faster. Writes are initially dumped to this area and that data is later moved over to the TLC area. This happenes as it would in a normal write-back cache - either during idle states or once the cache becomes full, which is what would happen during a sustained maximum speed write operation that is larger than the cache capacity. Here is the net effect with the cache in use and also when the cache becomes full:

DSC04638.JPG

For most users, even the smallest cache capacity will be sufficient for the vast majority of typical use. Larger caches appear in larger capacities, further improving performance under periods of large write demand. Here's the full spread of cache sizes per capacity point:

DSC04639.JPG

So there you have it, Samsung's new TurboWrite technology in a nutshell. More to follow (along with a performance review coming in the next few days). Stay tuned!

Western Digital's SSHD Black magic revealed

Subject: Storage | May 14, 2013 - 05:53 PM |
Tagged: sshd, cache, western digital, Black SSHD, Hybrid Drive

The Tech Report sat down with Matt Rutledge, Vice President of Western Digital's client computing group to discuss the software behind their new HDDs with an SSD cache.  Sandisk will be providing the hardware and WD who will be providing the custom caching software which will not be coded into the hardware but will function at the driver level.  Matt mentioned that this software can also make use of the system's memory and incorporate it into the cache as well though it was not completely clear if there will be many user editable settings.  Check the interview out.

tr_black.jpg

"WD revealed that its hybrid drives will use SanDisk iSSD flash components. The announcement was devoid of details on how the caching system works, but we can now shed new light on the software-managed scheme."

Here are some more Storage reviews from around the web:

Storage

Source:

Seagate Introduces SSHD Lineup with Dual Mode NAND Cache

Subject: Storage | March 8, 2013 - 09:20 AM |
Tagged: sshd, solid state, Seagate, Intel SRT, cache, adaptive memory

Following the announcement that the company would be axing 7200 rpm notebook drives, Seagate has introduced its third generation hybrid hard drives. The new Seagate Solid State Hybrid Drives (SSHD) will initially launch with two notebook drives and a single desktop-sized drive. The hybrid drives will combine a spinning platter drive with 8GB of NAND flash with Seagate’s Adaptive Memory tech that will reportedly cache reads as well as writes.

The 2.5” notebook SSHDs include a 7mm model that combines 500GB of mechanical storage and 8GB of Adaptive Memory cache. This model will retail for around $80. There will also be a slightly larger 9.5mm  with 8GB of cache and 1TB mechanical hard drive capacity. The 1TB model utilizes two 500GB, 5400RPM platters and will retail for just under $100.

Seagate SSHD.jpg

The desktop SSHDs come in 3.5” form factor and will initially use 7200 RPM platters.  Seagate will offer up to 2TB of mechanical storage with its SSHDs and 8GB of NAND flash for caching.  Seagate claims that its desktop SSHD is up to four times faster than other mechanical hard drives, (according to PC Mark Vantage) which is likely due to the Adaptive Memory technology caching frequently used data on the flash memory and the use of 1TB platters. The 1TB and 2TB SSHD will cost around $100 and $150 respectively. Naturally, the SSHDs will carry a small premium over traditional mechanical hard drives. They will still be much more price-efficient than Solid State Drives for the storage offered (though I would still like to see a larger NAND cache).

Interestingly, Tech Report was able to glean a few more details about Seagate’s third generation hybrid drives. Reportedly, the drives will be capable of writing as well as reading to/from the NAND cache. That is a major step up from previous generation’s which limited the drive’s flash storage to a read-only cache. Seagate has reportedly built the drives such that they will have enough capacitance to flush the write cache in the event of a power failure (so that you will not lose any data). The dual mode NAND term stems from Seagate’s ability to use SLC for boot data and the write cache and address the remaining NAND as MLC flash. Unfortunately, details are scarce on how Seagate is doing this.

The SSHDs will come with three year warranties, but Seagate has rated the NAND flash at a lifespan of at least five years. In an neat twist, Seagate is also allegedly working on another SSHD implementation that will combine a mechanical hard drive and a larger NAND cache. However, the flash memory will be managed by Intel’s Smart Response Technology instead of Seagate’s own Adaptive Memory tech (which doesn't need additional drives, unlike SRT). Using the port multiplexing aspect of the SATA spec, Seagate will be able to put both drives into a single 3.5” form factor hybrid drive. Admittedly, this is the Seagate SSHD that I am most excited about, despite the fact that it’s also the drive I know the least about. I’m interested to see what kind of performance Seagate can wring out of the larger cache!

Source: Seagate