Virtualization is nothing new. Anyone with a RAID array has done virtualization. In the 90's, virtualization of storage began to get interesting, but suffered from a lack of robust tools. In reflex, companies in the virtualization business begged not to have the V-word mentioned as part of their coverage in the trade press.
But now, virtualization is found under every infrastructure. Virtualization at the server level is an increasingly popular tool to consolidate servers and manage the increasingly large flow of data. Of course, there is a consequence: latency comes into play when you add another layer on the stack. But at current processing speeds, the latency is less and less a consequence.
The next logical step is to virtualize storage...not as easy as one would think. There is a scattering of proprietary solutions, stronger and weaker, to realize the dream of true any-to-any storage. Fine as this is, and useful as this kind of connectivity certainly is, the growth of the data center is not usually tied to a single vendor, a single file scheme, a single operating system. I have to wonder whether storage virtualization is useful anywhere except in one-vendor shops.
Additionally, storage virtualization may not be for everyone. A balancing test between the cost of adding storage devices and the expense of the virtualization, license and warranty needs to be made. An installation may be too small, and scaleability needs to be considered.
But virtualization is a popular bandwagon, now, and big names are jumping. Just let the buyer make an informed buying decision.
Thursday, April 24, 2008
Friday, February 8, 2008
The Preannounced Death of Tape
I'm doing series of articles on tape technology, and during the research phase found more than a little of the "tape is dead" rhetoric. When you hang around storage journalism, you hear the death knell for tape at least once a year. One commentator referred to tape as "yesterday's technology." I'd dismiss it for the hype that it is, but too frequently it has the feel of the new innovator dismissing the older observer as hopelessly old-fashioned. The first message should be, then: don't change your existing tape solutions just to be trendy, or to be considered "up to date." Being up to date with storage can last a few months at best. In 1952, the RAMAC from IBM held 5 megabytes...and had this suspect thing called "random access." United Airlines didn't buy that early RAMAC to be trendy...they bought it to be useful. For the first time, you could get flight information quickly and conveniently.
You don't buy or keep tape solutions to be trendy, you buy or keep them to be useful. Storage administrators understand that much corporate data is only accessed infrequently. Yet storage volumes are multiplying each year as corporations generate more electronic content and struggle to meet complex business transactions, regulatory compliance mandates, and e-discovery requirements. Expanding storage on the storage area network (SAN) simply by adding expensive fiber channel disks is typically not a cost-efficient option.
Tape cannot be dismissed as yesterday's technology if it solves today's business problems. Tape subsystems conserve power in the power-stressed data center. It is removeable to other sites, protecting essential information in case of a natural or man-made cataclysm. (The buzzwords are business continuity and disaster recovery.) Data centers can scale with extra tape cartridges rather than more drives, in most cases. Tape's cost continues to be competitive.
Just because a technology stands the test of time doesn't make them timed out. Like the old Monty Python sketch where the death carts were trying to collect a body during the Black Death: "I'm not dead yet." Neither is tape in the data center.
You don't buy or keep tape solutions to be trendy, you buy or keep them to be useful. Storage administrators understand that much corporate data is only accessed infrequently. Yet storage volumes are multiplying each year as corporations generate more electronic content and struggle to meet complex business transactions, regulatory compliance mandates, and e-discovery requirements. Expanding storage on the storage area network (SAN) simply by adding expensive fiber channel disks is typically not a cost-efficient option.
Tape cannot be dismissed as yesterday's technology if it solves today's business problems. Tape subsystems conserve power in the power-stressed data center. It is removeable to other sites, protecting essential information in case of a natural or man-made cataclysm. (The buzzwords are business continuity and disaster recovery.) Data centers can scale with extra tape cartridges rather than more drives, in most cases. Tape's cost continues to be competitive.
Just because a technology stands the test of time doesn't make them timed out. Like the old Monty Python sketch where the death carts were trying to collect a body during the Black Death: "I'm not dead yet." Neither is tape in the data center.
Copyright in Cyberspace
Mind if I reprint something? I was on the phone last night with a friend who had a problem with the music download rules. Internet info should be free, he said. I looked at the issue way back in 2000 while I was still with Computer Technology Review. I've seen no reason to change my view. Here it is:
I read two articles over the weekend. They were completely at odds. One was in an Internet-specific journal, suggesting that the Internet mandates new thinking about law, commerce, technology, and culture. The other was in my metropolitan newspaper and it was basically a call to arms, boldly declaring that property is property, regardless of the Web. The status of ideas and creative verve is central to the function of the systems integrator and the OEM. Therefore, the issue of copyright is relevant to your work.
Right now, two major technology issues are very much in the news. One is the Solomon-like decision to split Microsoft into two non-cooperating entities. The other is the war over music copyrights. The war begs a number of questions about the nature of ideas and how or whether they are shared. It is this latter issue that needs some exploring.
On one side, cyber-protectors insist that the Web is a different breed of cat. They cultivate a world view that says that no one can clearly define what copyright and intellectual property mean in virtual space. Those people who haunt the Web have built their own cultural databases, if you will. They are made up of Web bookmarks, movies, music, trendy home pages, and more. They represent a point of view well expressed by Lawrence Lessig in Code And Other Laws Of Cyberspace: "If you 'take' my idea, I still have it. If I tell you an idea, you have not deprived me of it. Your consumption does not lessen mine. Ideas, at their core, can be shared with no reduction in the amount the 'owner' can consume. The difference is fundamental..." Finally, they suggest that the structure of the Internet alters the way that one must view many traditional ethics and values, including definitions of theft, content, and property.
On the other hand, my newspaper reprinted a "Perspective" piece based on the speech of a music industry executive. The content was what you would expect from the business community. Intellectual property is property, period. Ideas are property and, if intellectual property is not protected in every situation, it will face extinction. References are made to the My.MP3.com case in New York and the Napster case in Northern California. This thinking also condemns anonymity on the Internet as the digital equivalent of wearing a ski mask while robbing a bank. The implication is that privacy and anonymity are blurred in the speeds and feeds of cyberspace and the lack of distinction would harm our culture, our society, and the long-term growth of the Internet.
The truth, if there is a truth, lies somewhere between the free-wheeling views of cyberspace dwellers and the narrow, profit-inspired views of those who exploit ideas and creativity for sometimes outrageous profit. Many of the original settlers in the New World that became the United States came here to establish rights to property ownership. Such rights were denied by many European nations due to religious affiliation or political alignment. Nothing is closer to the American idea of freedom than the ability to enjoy the fruits of one's labor.
Yet even America's founding fathers were not united in their views on the issue of exploiting ideas for gain. Thomas Jefferson might have been in the middle of the cyberspace/copyright dispute when he wrote: "That ideas should be freely spread from one to another over the globe, for the moral and mutual instruction of man and improvement of his condition seems to have been peculiarly and benevolently designed by nature when she made them like fire, expansible over all space without lessening their density at any point, and, like the air,... incapable of confinement or exclusive appropriation. Inventions, then, cannot in nature be a subject of property."
Music is a particularly difficult area of intellectual property. It is a multibillion-dollar industry and dynasties can rise and fall based on ownership of a songwriter's vision or a performer's interpretation, but music is more than a cold piece of property, like a telephone or an automobile. Music has a cultural presence, impacting listeners whether they are consumers of the music's medium or not. There is likely some point at which music that molds our culture belongs "to the ages." It is the province of philosophers, musicians, and legislators to identify such a point. It is too important an issue to leave to the business community. Intellectual property is property...but it is often much more and this added dimension is what makes music a difficult area of IP.
In information technology, the same kind of impact isn't always obtained. The newest processor or operating system is unlikely to excite the same kind of cultural tidal wave as the newest musical release from a popular rock group, but the theory is the same. Are the copyright laws, as currently instituted and enforced, protecting the users of intellectual properties, as well as their owners?
The assault made by music industry executives and their legal attack dogs on anonymity on the Web deserves comment. They suggest that we must restrict the anonymity behind which people hide to commit crimes and that anonymity must not be equated with privacy. I believe the logic is flawed from both legal and historical perspectives. It is the same logic that bigoted legislators, especially in the deep South, used to argue that membership lists for the NAACP or the Congress on Racial Equality should be made generally available. Anonymity is valid to protect a user from embarrassment or to shield a user who elects to speak out on a controversial issue. (Check out the court opinion in ACLU vs. Miller in 1997 down Georgia way.) Wiping out this protection of free speech and association could be considered too great a sacrifice. At the beginning of this editorial, the analogy was used likening anonymity to wearing a ski mask to rob a bank. That does not mean that you should not be making ski masks. They have their place: on the slopes or on a public street, if the user wants.
If we were living in Voltaire's "best of all possible worlds," or Shangri-la, or Erewhon, or Thomas More's Utopia, the views of the cyber-guys and cybergals would probably hold forth and all ideas would be freely flowing. If we were in the Renaissance, the brilliant composers, performers, artists, and inventors would have patrons who would support them as they provided creative product for the betterment of the human condition and the development of cultures.
Yet we have very few modern-day Medicis and, for the time being, the composers, performers, artists, and inventors need to exploit their creativity financially. This is tremendously sad... I would prefer to live in the Utopia or Shangri-la, but creativity could suffer if there is no recompense for creativity and originality. Anthony Trollope summarized it well in his An Autobiography: "Take away from English authors their copyrights and you would very soon take away from England her authors." Yet true as this is, the current copyright laws are inadequate because they are focused on property rights alone. Those we trust with legislating such issues must make some kind of provision for culture: "the moral and mutual instruction of man and improvement of his condition."
I read two articles over the weekend. They were completely at odds. One was in an Internet-specific journal, suggesting that the Internet mandates new thinking about law, commerce, technology, and culture. The other was in my metropolitan newspaper and it was basically a call to arms, boldly declaring that property is property, regardless of the Web. The status of ideas and creative verve is central to the function of the systems integrator and the OEM. Therefore, the issue of copyright is relevant to your work.
Right now, two major technology issues are very much in the news. One is the Solomon-like decision to split Microsoft into two non-cooperating entities. The other is the war over music copyrights. The war begs a number of questions about the nature of ideas and how or whether they are shared. It is this latter issue that needs some exploring.
On one side, cyber-protectors insist that the Web is a different breed of cat. They cultivate a world view that says that no one can clearly define what copyright and intellectual property mean in virtual space. Those people who haunt the Web have built their own cultural databases, if you will. They are made up of Web bookmarks, movies, music, trendy home pages, and more. They represent a point of view well expressed by Lawrence Lessig in Code And Other Laws Of Cyberspace: "If you 'take' my idea, I still have it. If I tell you an idea, you have not deprived me of it. Your consumption does not lessen mine. Ideas, at their core, can be shared with no reduction in the amount the 'owner' can consume. The difference is fundamental..." Finally, they suggest that the structure of the Internet alters the way that one must view many traditional ethics and values, including definitions of theft, content, and property.
On the other hand, my newspaper reprinted a "Perspective" piece based on the speech of a music industry executive. The content was what you would expect from the business community. Intellectual property is property, period. Ideas are property and, if intellectual property is not protected in every situation, it will face extinction. References are made to the My.MP3.com case in New York and the Napster case in Northern California. This thinking also condemns anonymity on the Internet as the digital equivalent of wearing a ski mask while robbing a bank. The implication is that privacy and anonymity are blurred in the speeds and feeds of cyberspace and the lack of distinction would harm our culture, our society, and the long-term growth of the Internet.
The truth, if there is a truth, lies somewhere between the free-wheeling views of cyberspace dwellers and the narrow, profit-inspired views of those who exploit ideas and creativity for sometimes outrageous profit. Many of the original settlers in the New World that became the United States came here to establish rights to property ownership. Such rights were denied by many European nations due to religious affiliation or political alignment. Nothing is closer to the American idea of freedom than the ability to enjoy the fruits of one's labor.
Yet even America's founding fathers were not united in their views on the issue of exploiting ideas for gain. Thomas Jefferson might have been in the middle of the cyberspace/copyright dispute when he wrote: "That ideas should be freely spread from one to another over the globe, for the moral and mutual instruction of man and improvement of his condition seems to have been peculiarly and benevolently designed by nature when she made them like fire, expansible over all space without lessening their density at any point, and, like the air,... incapable of confinement or exclusive appropriation. Inventions, then, cannot in nature be a subject of property."
Music is a particularly difficult area of intellectual property. It is a multibillion-dollar industry and dynasties can rise and fall based on ownership of a songwriter's vision or a performer's interpretation, but music is more than a cold piece of property, like a telephone or an automobile. Music has a cultural presence, impacting listeners whether they are consumers of the music's medium or not. There is likely some point at which music that molds our culture belongs "to the ages." It is the province of philosophers, musicians, and legislators to identify such a point. It is too important an issue to leave to the business community. Intellectual property is property...but it is often much more and this added dimension is what makes music a difficult area of IP.
In information technology, the same kind of impact isn't always obtained. The newest processor or operating system is unlikely to excite the same kind of cultural tidal wave as the newest musical release from a popular rock group, but the theory is the same. Are the copyright laws, as currently instituted and enforced, protecting the users of intellectual properties, as well as their owners?
The assault made by music industry executives and their legal attack dogs on anonymity on the Web deserves comment. They suggest that we must restrict the anonymity behind which people hide to commit crimes and that anonymity must not be equated with privacy. I believe the logic is flawed from both legal and historical perspectives. It is the same logic that bigoted legislators, especially in the deep South, used to argue that membership lists for the NAACP or the Congress on Racial Equality should be made generally available. Anonymity is valid to protect a user from embarrassment or to shield a user who elects to speak out on a controversial issue. (Check out the court opinion in ACLU vs. Miller in 1997 down Georgia way.) Wiping out this protection of free speech and association could be considered too great a sacrifice. At the beginning of this editorial, the analogy was used likening anonymity to wearing a ski mask to rob a bank. That does not mean that you should not be making ski masks. They have their place: on the slopes or on a public street, if the user wants.
If we were living in Voltaire's "best of all possible worlds," or Shangri-la, or Erewhon, or Thomas More's Utopia, the views of the cyber-guys and cybergals would probably hold forth and all ideas would be freely flowing. If we were in the Renaissance, the brilliant composers, performers, artists, and inventors would have patrons who would support them as they provided creative product for the betterment of the human condition and the development of cultures.
Yet we have very few modern-day Medicis and, for the time being, the composers, performers, artists, and inventors need to exploit their creativity financially. This is tremendously sad... I would prefer to live in the Utopia or Shangri-la, but creativity could suffer if there is no recompense for creativity and originality. Anthony Trollope summarized it well in his An Autobiography: "Take away from English authors their copyrights and you would very soon take away from England her authors." Yet true as this is, the current copyright laws are inadequate because they are focused on property rights alone. Those we trust with legislating such issues must make some kind of provision for culture: "the moral and mutual instruction of man and improvement of his condition."
Tuesday, January 15, 2008
Happy New Year and Key Management
First and foremost, Happy 2008 to all. Hope it is profitable and satisfying in many, many ways.
A comment on my NeoScale blog from Jon Toigo (whose own blog drunkendata.com never fails to amuse, challenge or provoke) asked about my take on key management where all the discussion is on encryption.
Key management is fundimental to the management of encrypted data. Encryption is, of course, very much in the news since information "in clear" has been so much at risk. The most recent report is the IRS managed to lose unencrypted tapes of part of their database. They no longer use tape, but a secure PDF. How easy it would have been to shoulder the small performance penalty and encrypt...the lack of a key would have rendered the tapes useless to any except someone with a usable key.
Key management facilitates authorization, denial of authorization, key tracking, key ownership and other information essential to securing stored data. There is some question on where the key management should be placed. Storage encryption is typically a private key creature, and some recommend on storing the key at a closest possible point to the data. Some, as NeoScale did, used an appliance in the data path to accomplish key management.
My only problem with the variety of key management solutions is that you can't leverage your one solution for multiple kinds of encryption at multiple locations. When a sensible solution for centralized key management comes up, that will really be good news. The management of data is fully as important as the repository for it, and key management is just another element of overall data management. But the proprietary nature of private key encryption will continue to require multiple key management tools for some time to come.
A comment on my NeoScale blog from Jon Toigo (whose own blog drunkendata.com never fails to amuse, challenge or provoke) asked about my take on key management where all the discussion is on encryption.
Key management is fundimental to the management of encrypted data. Encryption is, of course, very much in the news since information "in clear" has been so much at risk. The most recent report is the IRS managed to lose unencrypted tapes of part of their database. They no longer use tape, but a secure PDF. How easy it would have been to shoulder the small performance penalty and encrypt...the lack of a key would have rendered the tapes useless to any except someone with a usable key.
Key management facilitates authorization, denial of authorization, key tracking, key ownership and other information essential to securing stored data. There is some question on where the key management should be placed. Storage encryption is typically a private key creature, and some recommend on storing the key at a closest possible point to the data. Some, as NeoScale did, used an appliance in the data path to accomplish key management.
My only problem with the variety of key management solutions is that you can't leverage your one solution for multiple kinds of encryption at multiple locations. When a sensible solution for centralized key management comes up, that will really be good news. The management of data is fully as important as the repository for it, and key management is just another element of overall data management. But the proprietary nature of private key encryption will continue to require multiple key management tools for some time to come.
Tuesday, December 18, 2007
The Real Cautionary Tale
The mainstream trade press is using the case of the destroyed CIA tapes as a cautionary tale in preservation of tapes for businesses and the courts. But what I see as a great danger is that IT managers could get caught in the web of spoliation of evidence, and 'I was just following orders' is not going to cut it.
Lawyers and courts use the term spoliation in state and Federal procedure to refer to the withholding, hiding, or destruction of evidence relevant to a legal proceeding. This is a violation of the criminal law and is not new. Spoliation has two consequences: first the act may result in fines and jail time for the parties who engaged in the spoliation. Additionally, case law added a particularly nasty curve ball called the 'spoliation inference.' The spoliation inference is a negative evidentiary inference that a judge or jury can draw from a party's destruction of a document or thing that is relevant to an ongoing or reasonably foreseeable civil or criminal proceeding: The 'finder of fact' can review all evidence uncovered in as strong a light as possible against the spoliator. There is no benefit of the doubt here.
For the IT manager, who is now very much an evidence manager at law, destruction of data must be planned and carried out according to established policies and procedures that can rebut the claim of spoliation. Destruction of data on an ad-hoc basis sews the seeds of litigation disaster.
We sometimes like to make fun of CEOs, saying their priorities are to make money, to save money, and to stay out of little orange jumpsuits. IT management priorities now have to include this last, sad priority.
Lawyers and courts use the term spoliation in state and Federal procedure to refer to the withholding, hiding, or destruction of evidence relevant to a legal proceeding. This is a violation of the criminal law and is not new. Spoliation has two consequences: first the act may result in fines and jail time for the parties who engaged in the spoliation. Additionally, case law added a particularly nasty curve ball called the 'spoliation inference.' The spoliation inference is a negative evidentiary inference that a judge or jury can draw from a party's destruction of a document or thing that is relevant to an ongoing or reasonably foreseeable civil or criminal proceeding: The 'finder of fact' can review all evidence uncovered in as strong a light as possible against the spoliator. There is no benefit of the doubt here.
For the IT manager, who is now very much an evidence manager at law, destruction of data must be planned and carried out according to established policies and procedures that can rebut the claim of spoliation. Destruction of data on an ad-hoc basis sews the seeds of litigation disaster.
We sometimes like to make fun of CEOs, saying their priorities are to make money, to save money, and to stay out of little orange jumpsuits. IT management priorities now have to include this last, sad priority.
Monday, November 26, 2007
NeoScale a Hot Prospect
Had a note from my friend Barbara Nelson over at NeoScale. NeoScale does encryption and key management using an appliance in the storage infrastructure. Barbara relates that, just as the company was readjusting to focus more on the key management product, multiple bids were appearing for the company.
NeoScale is right on that the best way to assure value from encryption is to enjoy comprehensive key management. Obviously, others see this advantage. I'm looking forward to seeing how Barbara and her board navigate this 'hot prospect' status. Quantum never did understand what an asset they had in Barbara, who is one of the few executives I know who has been able to so effectively blend engineering acumen and business sense. But they let her get away to realize her dream of running a company. Whoever acquires NeoScale should be very conscious of everything that Barbara brings to the party.
NeoScale is right on that the best way to assure value from encryption is to enjoy comprehensive key management. Obviously, others see this advantage. I'm looking forward to seeing how Barbara and her board navigate this 'hot prospect' status. Quantum never did understand what an asset they had in Barbara, who is one of the few executives I know who has been able to so effectively blend engineering acumen and business sense. But they let her get away to realize her dream of running a company. Whoever acquires NeoScale should be very conscious of everything that Barbara brings to the party.
Thursday, November 15, 2007
Slow Students of Archiving
Brian Fonseca at Computerworld reports that the mayor of Washington has abandoned an email retention policy that would have automatically and permanently erased government emails after a six month term. This announcement took place at about the same time that AP reported that US Justice H. Kennedy ordered the White House to preserve copies of all of its emails in an ongoing disagreement with the Archivist of the United States.
Issues of archiving are commonplace in the trade and now the mainstream press. There are many reasons for public officials to want to do extensive deletion of files, some legit and some less so. Emails are now business records, as important as any contract or warranty. They have evidentiary value in courts of law all over the country. They keep businesses going in a way faxes and wires never could match. But they are also instruments that could seal a company's liability or a government agency's guilt or innocence. They also take up megabytes and megabytes of storage.
As far as the government battles are concerned, I simply see them as slow students. Archiving is a fact of life, and anyone who protests that it costs too much or takes too much staff time is howling in the wind. Business is seeing the inevitability because of regulatory mandates and the persistent fear of litigation that infects the nation. Trying to whitewash a political legacy, another purpose of deleting emails, is useless...the truth will ultimately out. No matter what you think of de-duplication, there is a copy out there somewhere.
With all of the excellent software out there for archiving purposes, lost emails will increasingly become suspect. Archiving and retrieval will become a mainstay of data management, and effective data management is the foundation of the efficient storage infrastructure. The most lively debate, however, is what kind of hardware and media is to act as our archiving technology of choice. Hard disk is doing its level best to take center stage in archiving, as it is in operational and transaction processing. Tape technology and optical technology also have strong cases to make that they are either the more economical approach or the approach that delivers the most longevity. Plasmon, a leader in optical technology, is preparing to introduce a new optical archiving product, and has named a new CEO to push the tech.
Some consider the primacy of hard disk a manifest destiny, due to a dropping price per megabyte and the performance delivered by spin physics. Some think tape and VTL are merely postponing the inevitable. What do you think?
By the way, have you ever noticed that the press coverage in the trades focus on the retention of these invaluable emails, but look much less frequently at their secure destruction?
Cheers.
Issues of archiving are commonplace in the trade and now the mainstream press. There are many reasons for public officials to want to do extensive deletion of files, some legit and some less so. Emails are now business records, as important as any contract or warranty. They have evidentiary value in courts of law all over the country. They keep businesses going in a way faxes and wires never could match. But they are also instruments that could seal a company's liability or a government agency's guilt or innocence. They also take up megabytes and megabytes of storage.
As far as the government battles are concerned, I simply see them as slow students. Archiving is a fact of life, and anyone who protests that it costs too much or takes too much staff time is howling in the wind. Business is seeing the inevitability because of regulatory mandates and the persistent fear of litigation that infects the nation. Trying to whitewash a political legacy, another purpose of deleting emails, is useless...the truth will ultimately out. No matter what you think of de-duplication, there is a copy out there somewhere.
With all of the excellent software out there for archiving purposes, lost emails will increasingly become suspect. Archiving and retrieval will become a mainstay of data management, and effective data management is the foundation of the efficient storage infrastructure. The most lively debate, however, is what kind of hardware and media is to act as our archiving technology of choice. Hard disk is doing its level best to take center stage in archiving, as it is in operational and transaction processing. Tape technology and optical technology also have strong cases to make that they are either the more economical approach or the approach that delivers the most longevity. Plasmon, a leader in optical technology, is preparing to introduce a new optical archiving product, and has named a new CEO to push the tech.
Some consider the primacy of hard disk a manifest destiny, due to a dropping price per megabyte and the performance delivered by spin physics. Some think tape and VTL are merely postponing the inevitable. What do you think?
By the way, have you ever noticed that the press coverage in the trades focus on the retention of these invaluable emails, but look much less frequently at their secure destruction?
Cheers.
Subscribe to:
Posts (Atom)