Installers, integrators and end users should have a sense of hard drive projected mean time to failure and be able to broadly assess failure-free operating periods.
Modern hard disk failures are usually related to the head-disk interface and there are a range of issues that can impact on this. First up there’s handling damage, then temporary interface disruption, media damage and thermo-mechanical stability of the read-write structures. Depending on application, environment and the ‘lifestyle’ of the drive, the causes may relate to the customer end. In this case, damage will be caused by mechanical shocks, extrinsic contamination and condensation.
If the cause of the problem relates to manufacture, the problem may not necessarily be one of design but could be caused be induced during the manufacturing process itself. Such issues will include intrinsic contamination, servo errors, particulates, increases in lube thickness, localised weak or thing regions in the carbon overcoat or by head contacts with the disk. Given HDDs have integral logic boards, other issues might be solder joint fatigue or dendritic growth of solder that could lead to shorts in extreme cases.
For the security manager and the tech many of the whys and wherefores are going to be moot. Obviously you should push for the installation of more reliable constant duty HDDs but once they’re installed the key contribution you can make is limited to maintenance based on understanding mean time to failure and mapping out a realistic failure-free operating period.
How long will your hard disk last and what sort of disk should you use given the importance of HDD performance in terms of your CCTV system’s overall reliability? One thing is certain. Hard disk reliability has come a long way in 25 years. When those early American Dynamics’ Intellex units hit the market in Australia in 1997, mean time between failure was up from 150,000 hour average of 1993 to around 250,000. By 1999 MTTF was around 400,000 hours. And by 2014, numbers for the best HDDs had reached 1 million hours – a lifespan of up 110 years, depending on usage. And in 2021, leading HDDs like Seagate’s Exos Enterprise and Western Digital’s WD Gold have MTTFs of 2.5 million hours – once unthinkable performance.
Taking MTTF into account is important for security managers and maintenance technicians alike. But claimed MTTF is not a guarantee once you’re outside the typical 12-month warranty period and not all HDDs are created equal, according to recent data. This applies especially for HDDs used in constant duty applications.
When specifying and building a system you need to balance out the cost of higher quality HDDs with the cost of replacing drives at shorter intervals. Given the cost of maintenance, especially with smaller systems, the best economy of scale will almost always be with the higher quality, better performing solution, given the cost variation at purchase point may be less than $50.
From a maintenance procedure point of view, the key number to pay attention to is the failure-free operating period. The FFOP concept is the point in time after activation in a system that HDD mechanisms will begin to wear out. Once that FFOP is established you can think about maintenance management and performance screening checks that might prevent early failure. Such screens might simply be measuring average temperature HDDs are subject to, testing dust and moisture levels, taking in account vibration and either winding in the expected FFOP or remedying problems to ensure expected longevity targets are met.
From a technical point of view, HDDs have delicate or sensitive design elements or components that can be thought of as wear-out mechanisms. Different applications will challenge different wear-out mechanisms, whether these be the head-disk interface or PZT actuators. You need to be aware that manufacturers often base their assumptions of FFOP on tests that bring such components to failure and you should establish what the parameters of those tests was.
There’s no point relying on predictions of 6 years FFOP if the study is based on PZT actuator failure under conditions of zero vibration in an ambient temperature of 25 degrees C and your application calls for a mobile DVRs in an environment with part time air conditioning, variable humidity and lashings of vibration.
Different manufacturers have different rules when it comes to establishing reliability. They also set down different temperature tolerances and as a general rule don’t take into account variations like vibration or extreme temperatures. That means you need to consider these elements yourself. Experience will play a major part. Once you’ve been using an HDD in a given application for a couple of years you’re going to get a feel for the sort of performance a particular product is likely to deliver.
A key issue for some video surveillance recording applications is the level of active operation and this may vary depending on the nature of the recording system you have installed. If constant duty is required there will be pressure on elements of the drive and greater need for adequate ventilation or cooling. At the same time, a drive that is used 80 per cent of the time as might be the case in a RAID 5 application where elements of the video stream are stored as strips of data across 5 locations on different drives, will suffer wear in other ways.
Drive spindles can cop a beating when the drive is momentarily powered down in the stop/start course of recording parts of events. Some drive manufacturers only project a 50,000 cycle limit for start/stop cycles and that may be a major problem if you’re application relies on start/stop cycles to get the job done. Consider that over a 5-7 years lifespan an HDD may only start and stop 25 times per day or there will be an impact on reliability in the long term. Consider eco-friendly drives with a pragmatic eye. These spin down when not in use and restart when needed and that effort puts miles onto the hardware.
Take-off and landing repetitions are a leading cause of failure though the problems vary based on design. Consider that when a hard drive is powered down, the springs or actuator coil (depending on the type of actuator) attached to the heads pull the head into the platter. This is called a landing. Every drive is designed to handle thousands of takeoffs and landings but since the head actually hits the platter, its best to have this happen on a section of platter where there is no data.
In a voice coil HDD design, the actuator coil springs the heads into a landing zone and lock position before the drive even stops spinning. The landing zone typically lies on the innermost cylinder or the outermost cylinder. This assures that the heads are not just let go of and left to drag along the platter until the platter stops, a problem common to the stepper motor design. When powered on, the drive automatically unparks itself and the coil is overcome by the magnetic force.
Another area of potential weakness and failure is the logic board – a PCB located underneath the drive. This board controls the spindle and head actuator and also translates data to a form usable by the controller and the rest of the system. Sometimes, an apparent disk failure is actually a failure of this logic board. In such a case, you can replace the logic board and regain access to the data held up on the drive by unplugging the board (and its attachment screws) and plugging a new board into the drive.
As we’ve touched mentioned before, there are many environments that are likely to challenge HDDs in electronic security application. These can occur notwithstanding the propensity for security operations to be left ‘sucking the hind tit’ when it comes to resources and infrastructure. Whether this means your designated node zero is an ill-constructed guard house with no air conditioning or the damp basement precinct of a railway station or shopping centre, the results are the same.
The reality of the lowly place security departments occupy in the pecking order means you need to find out what’s required to maximise HDD lifespan and take what measures you can to improve reliability, even if that only means moving racks away from wall (if you have the real estate) to allow better air flow. Better yet you might find a spot in a remote rack in your site’s computer room, though this will kick up thorny issues relating to physical security and access control.
Security managers, system designers and techs should favour HDDs built specifically for their sort of application. You want some tolerance for tough environments, integration of fly-by-wire technology, additional ventilation, as well as self-monitoring, analysis and reporting technology built into your hardware and your security applications. SMART technology allows the drive to monitor its functionality against alert thresholds. Regular checking of SMART status can pay off if your application cannot afford failure but was cramped by lack of resources and lacks redundancy.
#sen.news
