The technology, use cases and quality have improved over time
But moving onto an all-flash array is but one part of the equation
THE storage industry is booming, with venture capitalists chomping at the bit to invest in flash. The industry is seeing unprecedented hyper-growth, some in excess of 700% year-on-year.
IDC analysts have forecast that by 2015, the all-flash storage array market will reach US$1.2 billion in revenue, with an anticipated 58.5% CAGR (compound annual growth rate) over the forecast period.
Not bad for an industry that didn’t really exist five years ago. But what’s all the fuss about flash, and how can it work for your business?
Quick history lesson: Flash memory was named just that because its read/ write speeds reminded its inventors of the flash from a camera. Known as a solid-state storage device (SSD), flash has no moving parts, which allows it to be quicker, quieter, and more resistant to physical shock.
On the more technical side, it’s actually a type of non-volatile memory based on the logical circuit called a NAND ((Negated AND or NOT AND) gate. Non-volatile means that the memory cell retains its data even when the circuit receives no power.
Flash memory comes packaged either as SSDs, packaged standalone ICs (integrated circuits), or even as bare dies.
Storage suppliers that offer flash-based storage for enterprise applications are either going to either integrate SSDs into shelves, or place flash ICs onto a card with a serial interface such as PCI-Express.
While understanding the issues around SSDs can be tricky, the technology, use cases and quality have improved over time. As such some of the old truisms no longer apply.
Below are some of the terms and concepts specific to flash that are well worth absorbing:
A lot of people don’t know it, but flash cells lose their charge over time. They are also subject to wear and tear as you erase and re-write to them.
For enterprise storage, suppliers should be able to explain what they do to refresh data over time to account for the loss of charge to the flash cells. Whether using SSDs or packaged chips, some explanation of error detection and correction (called ECC) should be in the conversation.
Endurance or P/E cycles
P/E cycles mean program/ erase cycles, also referred to as endurance: The shelf life of a flash cell.
Flash cells get worn out as you erase and re-write to them, eventually resulting in device failure. The more you write to a cell, the weaker it becomes, gradually losing the ability to hold a charge when needed.
There are optimisation techniques to regulate the rate at which your flash wears out. Wear levelling, the process of distributing writes over flash blocks, is one of these.
Single, multi, and 3-bit level flash cells
Every flash cell has a ‘level’, or the logical method of defining the value of that stored bit. This then defines how many bits you can store in a flash cell, and it helps determine the cost per gigabyte at the device level.
As the industry tries to store more bits in a single cell, it is being presented with unique challenges. For one thing, while these newer semiconductor architectures are more efficient, they are also more complex.
Increased complexity also leads to an upward surge in error rates, which require different error connection techniques.
Single-level cell (SLC) devices can support P/E cycles in the tens of thousands, while multi-level cell (MLC) devices (two bits per cell), typically used for consumer applications, will range in the low to mid thousands of cycles, obviously failing sooner on average.
As such, SLC commands a considerable price premium over MLC, although the underlying circuit is identical. In a storage system, you should know which type of flash cell is being used.
Flash cells have to be erased before they can be written to. Comparatively, erasing takes a long time versus either a read or a write. This is because when you erase, you do it in larger chunks than when you write.
The result of this – and the fancy term is the asymmetric read/ write situation - is write amplification.
This means to write a page of data, first you have to erase the whole block. Valid pages are moved to another location, meaning the same data is written over and over again to different locations in the flash device, wearing out the flash cells faster.
One of the complexities of using flash is that reading and writing are asymmetric in size and duration. Erase, write and read operations all happen in different ways, concurrently, using shared routing resources.
When you combine asymmetric reads and writes, you can experience a significant variance in performance.
The big challenge for a storage system is to understand how your reads can be executed from a location that is pre-occupied with erasing and/ or writing.
When all is said and done, the bottom line is that managing the process of reading and writing data is absolutely imperative.
Moving onto an all-flash array is but one part of the equation, and when executed in combination with good software, only then can the full benefits of flash can be realised.
Michael Cornwell is chief technology officer, Asia Pacific and Japan, at Pure Storage. He holds 58 US patents awarded in flash and other storage technologies.
Data centres can go the SSD path without sacrificing existing investments
A budget-friendly approach to flash storage
All-flash arrays: Being a latecomer plays into XtremIO’s hand
New all-flash storage going to be game-changer: Dell
For more technology news and the latest updates, follow us on Twitter, LinkedIn or Like us on Facebook.