Does Copan’s demise portend the death of MAID? Marc Farley thinks MAID is toast and recently wrote:
I suspect MAID storage will quickly become an afterthought now, except for a small number of customers and applications that will keep the technology on life support. The problem with MAID is that there aren’t enough applications for selectively-spinning disks. Selectively spinning disk drives are more expensive than tape for archiving and are more problematic than standard disk systems for backup. That leaves applications such as video on demand, which is not a large enough market to float a serious startup these days. Thin provisioning for primary storage and dedupe for backup have become the technologies of choice for customers looking to increase the efficiency of storage.
Maybe…but I’m not so sure. First, for 3PAR (Marc’s company) MAID doesn’t make sense because it spreads data across drives and spinning down drives would create performance problems that would offset any benefits of spinning down devices. So 3PAR emphasizes higher utilization with thin provisioning. But that’s 3PAR’s angle…other companies including DataDirect Networks, EMC, Hitachi and Nexsan have implemented spindown in their arrays and in many cases the implementation is more flexible than Copan’s.
Originally as conceived by Copan, massive arrays of idle disks (MAID) were high-capacity, lower cost disk arrays for storing less active enterprise data and saving energy. The primary value proposition of MAID was lower operational costs, which stems from its capability to power down a portion of drives within the array, thereby lowering power consumption. Because spinning disk drives typically account for 80% of a storage array’s power consumption, MAID is, in concept, an effective technology for greening storage. Moreover, 90% of organizational data that is more than three months old, 70%-80% of all data in the organization overall, is inactive or never accessed.
The basic concept Copan conceived was to group and store data based on access frequency, placing rarely accessed data on devices that are turned off. Early criticisms of MAID included concerns about shutting down enterprise disk drives, which unlike laptop devices are engineered to be always on. Mainstream manufacturers began to introduce MAID-like features into disk arrays, lending credibility to the concept, although as Marc points out, the number of applications that can benefit from MAID is limited, in part because applications are not MAID-aware and spinning down drives makes applications freak out (because when applications request data and don’t get a near immediate response they think something’s gone terribly wrong).
Nonetheless, spindown has its place. Companies like EMC, Hitachi and Nexsan have introduced intelligent MAID that automatically manages different gradations of spindown along a spectrum such as the following:
- Normal – for online active data – no power savings;
- Park – which parks the disk drive’s heads while the platters continue to spin – 20-24% power savings;
- Slow Down – a mode that decelerates the spin speed of the disk platters from 7200 RPMs to 3600 – ~50% power savings;
- Sleep – a standby mode – which can deliver up to 85% power savings;
- Off – shuts down the disk device – effectively 100% savings.
In practice, few vendors shut down disk devices because of the concerns cited earlier. But there is little question that in concept, spinning down devices that store infrequently accessed data makes arrays more efficient.
Bill Mottram, a former Copan insider, points out in his blog several factors that led to Copan’s downfall. In addition, it turns out Copan was a classic feature company. All Copan had really was MAID and energy efficiency was its primary value proposition– not a good place to be in 2009.
But looking forward, as organizations announce and ship automated tiering software that can push infrequently accessed data to fat SATA devices, parking heads, slowing down spin speeds and putting drives in sleep mode makes sense for some array architectures. It’s worked in laptops for years and there’s no reason that as the technology matures, spindown won’t exist as an invisible means of improving disk array efficiency.