Seagate VP for branded products group, Mark Whitby, walks us through the fascinating world of storage, warning us of the dangers of not producing enough data and introducing us to the concept of the Zettabyte. The world, he says, will produce 44 of these by 2020, which might not sound a lot until you consider that a Zettabyte is 1^21 bytes.
TRP Tell us a little bit more about Seagate
MW Seagate is a leading provider of hard drives and storage solutions. From the videos, music and documents we share with friends and family on social networks, to servers that form the backbone of enterprise data centers and cloud-based computing, to desktop and notebook computers that fuel our personal productivity, Seagate products help more people store, share and protect their valuable digital content.
Seagate offers one of the industry's broadest portfolio of hard disk drives, solid-state drives and solid-state hybrid drives. In addition, the company offers an extensive line of retail storage products for consumers and small businesses, along with data-recovery services for any brand of hard drive and digital media type.
The company developed the world's first 5.25-inch hard disk drive (HDD) back in 1980 and in March, 2013, Seagate also became the first storage manufacturer to ship two billion drives globally. Seagate is headquartered in Cupertino, California and employs more than 50,000 people around the world.
TRP Why should people care about storage?
MW Data has never been more important. As valuable as oil and just as difficult to mine, model and manage, data is swiftly becoming a vital asset to businesses the world over.
Companies large and small are taking their first steps in data analytics, keen to gain an insight into how their customers behave and so better position themselves in the market place. Although still in its infancy, analytics holds the potential to one day allow them to find solutions, sell more products and develop customer trust.
While most businesses have yet to determine how to get the most from their existing data, let alone understand the masses of unstructured data from outside their organisation, they do accept its potential. Tomorrow's competitive advantage may well be driven by the ability to quickly identify the right data, collect it, analyse it and act on it.
In order to amass this valuable digital repository, however, there must first be ready storage capacity. And in order to drive all possible value from that data, it must also be stored in such a way as to be quick to access, efficient to manage, and low-cost to maintain. Unfortunately, therein lies the rub.
Data centres today are not equipped to be able to handle the anticipated influx generated by the Internet of Things, nor geared towards feeding it smoothly across to the analytics platforms where it can prove its worth. There is little chance that the billions of whirring silicon-based hard drives around the world will be able to keep up with the flood of data driven by the 26 billion connected devices (not including some 7.3 billion smartphones, tablets and PCs) that Gartner predicts will be in use by 2020.
TRP What do you think will be the main challenge facing the storage industry over the next 5 years?
MW Three words: data capacity gap.
We are entering a world where everything is connecting to everything else and the resulting big data is anticipated to solve virtually all our problems. However, by 2016, the hard drives housed in all those connected devices, whirring away in countless data centres, will start to reach their limits.
The total amount of digital data generated in 2013 was about 3.5 zettabytes (that's 35 with 20 zeros following). By 2020, we'll be producing, even at a conservative estimate, 44 zettabytes of data annually.
A zettabyte might not be a word you've heard of – even Word's spellchecker doesn't recognise it – but consider it in terms of a more familiar unit. A standard smartphone today will have around 32 gigabytes of memory. To get to one zettabyte you would have to completely fill the storage capacity of 34,359,738,368 smartphones.
At this current rate of production, by 2016 the world will be producing more digital information than it can easily store. By 2020, we can predict a minimum capacity gap of over six zettabytes - nearly double all the data that was produced in 2013.
TRP If the world is running out of storage, why can we not simply increase production of hard drives and build more data centres?
MW Unfortunately, the imminent breach between storage demand and production is not a problem that can so easily be solved. The fact of the matter is that it's far harder to manufacture capacity than it is to generate data. Building factory capacity that is capable of meeting such stratospheric demand would take hundreds of billions in investment. It's simply not a realistic option.
Another factor is the technology in use by the storage industry today. Even if the investment was there and thousands of new data centres could be commissioned, it's becoming more difficult on a molecular level to squeeze increasingly dense volumes of information onto the same amount of space.
Seagate produced its first ever hard drive in 1979: it had 5MB of storage and would have cost a few months' wages. What it could store today is about 2 seconds of low resolution video shot on a smartphone, or 2 high resolution photos. A modern 5TB hard drive will set you back less than £200 and is capable of storing 2 million photos, 2.5 million songs and about 1,000 movies. Although it's not physically any larger than our oldest hard drive, in capacity it's actually 1,000,000 times bigger.
So, while the ability to squeeze ever more dense data onto the same amount of space is a real testament to human ingenuity and engineering, it's starting to reach the point where new technologies will have to take over.
TRP What are some of the latest innovations in data storage that could help heal the data capacity gap in 2020?
MW Silicon may be the work-horse that has helped us get to where we are today, but it's starting to show its age. Fortunately, there is an impressive amount of innovation taking place in the industry at the moment and a number of these advances could help us to seal the data storage breach over the next five to 10 years.
RRAM (resistive random access memory) is one such example. A smart type of computer memory, this could, in theory, let us store tens or even hundreds of times as much data on our smartphone. The high difficulty and costs of production have meant that many companies have overlooked it in the past s, but researchers at Rice University have recently had a break-through. They have shown a way to make RRAM at room temperature and with far lower voltages. Some prototypes have even been proven to store data densely enough to enable a terabyte chip the size of a postage stamp.
If RRAM doesn't seem quite far enough removed from the world of silicon-based storage, there's also DNA to consider. Last year, a team of scientists from the European Bioinformatics Institute reportedly stored a complete set of Shakespeare's sonnets, a PDF of the first paper to describe DNA's double helix structure, a 26-second mp3 clip from Martin Luther King Jr.'s "I Have a Dream" speech, a text file of a compression algorithm, and a JPEG photograph in a strand of DNA, no bigger than a speck of dust. Another forward-looking team at Harvard University's Wyss Institute later brainstormed their way to successfully storing 5.5 petabytes, or 700 terabytes, of digital data into a single gram of DNA.
TRP Is Seagate developing any new storage solutions at the moment?
MW Heat-assisted magnetic recording (HAMR) is one new technology that Seagate is investing in. This method uses lasers to first heat the high-stability media before magnetically recording data. HAMR is expected to increase the limit of magnetic recording by more than a factor of 100 and this could theoretically result in storage capacities as great as 50 terabits per square inch - current hard drives generally have a capacity of a only few hundred gigabits per square inch. To put this in perspective, a digital library of all books written in the world would be approximately 400 TB— meaning that in the very near future conceivably all such books could be stored on as few as 20 HAMR drives.
While these technologies are still some way from our desks and data centres, these advances and others like them are certainly on their way. Innovation combined with the plunging cost of components is ultimately what's needed if we are to keep up with the world's growing demand for data storage.
TRP Will CIOs need to supplement existing storage resources?
MW CIOs certainly need to consider the implications of a data capacity gap for their business and address it by thinking strategically and longer term in regards to their storage resources.
Typical big data resides on traditional disk storage, using standard hardware and software components. Since companies began to store information, a large amount of standard infrastructure has built up around the process. Data centres today include legacy components comprised of hardware and software stack components. This approach is highly inefficient - in a single system there will often be several unsynchronised components caching the same data, each working independently, but with very poor results. In order for a company to get to a better cost and efficiency model, to match the requirements in the future, a better solution must be put in place.
One of the latest big data storage methods is a tiered model using existing technologies. This model utilises a more efficient capacity-tier based on pure object storage at the drive level. Above this sits a combination of high performance HDD (hard disk drives), SSHD (solid state hybrid) and SSD (solid state drives). SSHD hybrid technology such as this has been used successfully in laptops and desktop computers for years but today it is only just beginning to be considered for enterprise-scale data centres. This new method allows the most critical data to sit on the more expensive SSDs or hybrids, where it is easy and quick to access and well-placed to be processed by analytics platforms, while the less valuable meta-data sits on cheaper HDDs where it is still available and secure, but slower to access.
This potential part-solution to the data capacity gap is part of a growing trend for larger, more efficient data centres. That the world has grown from a planet producing just under one zettabyte per annum back in 2009, to potentially well over 44 in 2020, is truly astounding. Managing this—whether you're a technologist responsible for managing data, a business user who has the task of analysing it, or a consumer trying to manage the flood of your own digital information—will be an interesting challenge for all of us.
No comments:
Post a Comment