In December, we published a bulletin titled Cost of Exchange 2010 Storage Not An Issue. This said that storage is now so cheap that Exchange 2010 users needn’t worry about storage costs.

We still subscribe largely to that view. However, it’s not true for organizations that are committed to storage area networks (SANs). Here is a more detailed analysis.

SANs are used in many Exchange environments. They are expensive, however, so many Exchange users prefer to offload rarely accessed and archived material to less expensive storage.

For example, 18 terabytes (TB) of NetApp Fibre Channel SAN costs about $240,000 when switches, controllers, shelves, and software are included. After allowing for snapshots, this gives perhaps 12 TB of usable storage, which translates to $20 per gigabyte (GB). Most customers run Exchange 2010 in RAID 10, which doubles the cost. Furthermore, many customers have redundant sites to provide for better recoverability. So total SAN costs often reach an exorbitant $80 per GB.

With Exchange 2010, the archive is part of the main message store. Additionally, message stores now contain a lot of material formerly stored in Personal Storage Table (PST) files. If SANs are used, the cost of the additional storage is crippling: An extra 5 GB per user translates to $400 per user!

Exchange 2010 responds to the problem by supporting more modern, inexpensive storage. This also reduces message store I/O by some 90%, since Exchange 2003 provides for slower disk speeds. Modern non-SAN storage costs are about $0.30 per GB, including RAID 10 and redundant servers.

Microsoft hopes that the new drive technology will thus resolve the storage cost problem. It will be useful in many cases, but it is not a panacea. Many organizations have a sunk investment in SANs for a number of applications. These organizations will be resistant to using a separate set of technologies, with separate management tools, for the special case of email.

Thus, for Exchange users who have invested in SANs and are committed to the technology, storage cost is an issue for all versions of Exchange, including Exchange 2010. For SAN users, third-party archiving tools will continue to be an attractive way of allowing live mailboxes to live on SANs while archives reside on less expensive media.

… David Ferris, with particular thanks to Alan Elliot and Kevin Hood of Mirapoint, for educating me on the real costs

Hosted Exchange vendor Intermedia has added integrated telephony to its offering. The focus is on sales to SMBs, up to around 1,000 seats.

Thoughts:

• Hitherto the offering has consisted of email, IM, fax, and SharePoint.
• This is the first time we’ve heard of a hosted email offering providing integrated telephony.
• Telephony doesn’t require high bandwidth. However, it does require short latency. Packets need to get through reasonably quickly. Think of Skype’s periodic problems. Latency isn’t a problem if the packets go over an in-house network. But when they go over the public Internet, as they will with SMBs, quality of service becomes an issue. Time is on Intermedia’s side, and by 2020 latency shouldn’t be a problem for voice throughout the world.
• Intermedia is one of the few hosted Exchange vendors to be making money. The company is privately held and doesn’t disclose revenues, but we estimate its hosted Exchange revenues at around $40M annually.
• For more information, go here.

… David Ferris

This is the first of a series of bulletins on important features of Exchange archiving, namely: single instance storage, retention management, PST management, mailbox size, multi-mailbox search, and item-level restore. Beginning with Exchange 2010, many small and medium-size organizations may find that sufficient features are included, so that third-party archiving tools are no longer necessary.

One of the lesser-known changes to Exchange 2010 is the removal of single instance storage (SIS). The reason for this is related to an architectural change, disk I/O performance, and the availability of cheap disk.

There tends to be a trade-off between better disk I/O performance and reduced storage capacity. Architecturally, Exchange 2010 introduces a new per-mailbox table structure that replaces the original per-database table structure. The original per-database table structure was optimized for SIS, but disk I/O suffered. The new per-mailbox table structure improves disk I/O, but without SIS.

In place of SIS, Exchange 2010 uses compression. Only large, redundant attachments files truly benefit from SIS; otherwise, compression delivers roughly the same volume of data as SIS.

For further information, see this Exchange team blog discussion.

… Bob Spurzem

In addition to his role as Ferris analyst, Bob is director of product marketing at Permabit, which offers a grid–based disk storage system.

Hosted archiving vendor LiveOffice is expanding its supported data types. Hitherto, it has mainly archived Exchange email, with some instant messaging. As of March 2, LiveOffice is also archiving SharePoint teamspaces and Web site pages. The Web site support implies support for social networking Web sites such as LinkedIn, Facebook, and Twitter.

Costs:

• SharePoint archiving: $9.95/user/month
• Social Archive:$5/URL and $20/typical domain
• Volume discounts apply
• Search and e-discovery services are extra

For more information, go here.

Comments:

• Email is the most important data type to archive. However, ultimately what counts is information rather than the data types in which it’s stored. LO is right to extend its offering beyond email.
• SharePoint support is especially important. Very large amounts of compliance- and e-discovery-sensitive material are kept in these workspaces.
• The SharePoint support is initially limited to files and emails; e.g., no calendars or address books. Broader support for SharePoint content is planned.
• As of January 2010, FINRA started issuing guidance on social media. This is a clear indicator of the importance of these technologies to end customers.
• The Web site archiving works as follows: The user specifies the URLs of pages that should be archived, and how frequently a PDF snapshot should be stored. The PDFs are then archived; word lists generated by Adobe are indexed to provide for review and e-discovery.
• Most broker/dealers are small businesses, with simple Web sites. The static-URL-based scanning should suffice for most of them, as far as FINRA social networking compliance is concerned.
• The technology is very much version one in nature. It ignores the structure and meaning of the pages being stored, so searches are much weaker than they should be (more false positives and false negatives). Also, no doubt LO will want to add support for dynamically created URLs, such as database-driven content.
• APIs for social networks are still crude. They will improve, and that will help third parties such as LiveOffice provide much tighter integration.
• LiveOffice is privately held and doesn’t disclose its financials. Ferris estimates revenues at perhaps $30M annually for its hosted archiving and e-discovery business. The company is profitable.

… David Ferris

Last Friday (February 19), the chattering classes were gossiping about how cloud archiving vendor Smarsh had its Web site and phone lines down. Had the company suddenly gone down the toilet?

Everything was in fact OK. The company’s Ken Anderson told us:

• We had technical issues this morning that created some intermittent access to some of our services for some of our clients. No customer data has been lost or compromised – all back-end operations have been operating business as usual. Service has been restored, and steps are being taken to ensure that this sort of thing doesn’t happen again.
• The situation was exacerbated for us because it also affected our phones and Web site. Unfortunately, with the phone and Web site down, conditions were ripe for speculation. We have had an all-hands-on-deck client outreach effort today, making sure to communicate the status. (And as the communications rep, trust me, we are addressing these circumstances as well!)

So no real news here. It does illustrate, obviously, that cloud vendors need to have an extremely reliable offering. There is almost no tolerance for downtime.

… David Ferris

Iron Mountain announced it has acquired Mimosa Systems. This puts an end to over a year of rumors concerning Mimosa’s acquisition discussions with a number of possible suitors.

Iron Mountain has a vested interest in building bridges between customer on-premise deployments and its Digital Archive, and the company has been building partnerships with numerous on-premise archiving vendors to build bridges into their products. The idea is that as on-premise archives grow stale, customers will treat Digital Archive as a third tier of nearline storage at Iron Mountain’s hosting facilities. Today’s move places Iron Mountain into direct competition with other archiving vendors, and therefore casts a shadow over its partnership initiatives.

Thoughts:

• Allows Iron Mountain to offer on-premise archiving. That makes sense.
• Iron Mountain acquires a good-size customer pool that it will no doubt try to upsell into its Digital Archive.
• Also provides a new offering for Digital Archive customers.
• It seems rather late in the game to be entering the on-premise archiving market. Two powerful forces that Iron Mountain faces are the shift toward storing data in the cloud, and native archiving capabilities included in platforms such as Exchange 2010.
• By virtue of its leadership position in hardcopy archiving, Iron Mountain could be a dominant force in digital archiving. It has failed so far to achieve such a position.
• Iron Mountain is most reticent to give figures about its digital archiving business. We suspect that despite its major name in hardcopy archiving, those revenues are extremely modest.
• Iron Mountain has a series of archiving offerings that must now be sorted out. For example, it has its own code, plus offerings from MessageOne and MIMEcast. Products or services will need to be dropped, and interfaces made consistent. Integrating different products is difficult, and it’s not clear Iron Mountain has good skills in this regard.
• Iron Mountain has a very confusing mix of digital offerings. It badly needs to develop a clear vision of how these fit together.
• We hear that partly because of this confusion, the Iron Mountain sales team has a tough time selling the digital archiving portfolio.
• A million dollars here, a million dollars there, pretty soon you’re talking about real money ….. On the other hand, Iron Mountain is a large company. It can afford to invest $100M or so in acquisitions, without having to worry too much about the success of the investment.
• Iron Mountain has made a number of acquisitions, and established a number of technology partnerships. Some have gained traction, others haven’t. Overall, it feels like a strategy of trying this and that, and seeing what works. If you’re a big, rich, successful company, you can afford that approach.

Regarding the transaction:

• We estimate Mimosa was generating about $25M/year and had flat revenues. (ACTUALLY IT DID $20.6M IN CALENDAR 2009)
• Mimosa had never been profitable. We estimate it was losing about $10M/year. (ACTUALLY IT HAD $32.7M EXPENSES IN CALENDAR 2009, SO WAS LOSING ABOUT $12M/year)
• Mimosa took around $70M to $100M in investment funding. The investors had evidently decided not to put in more money, and the company was forced to find a buyer. See our December 2009 bulletin, “Mimosa Funding Challenges”.
• The $112M valuation is a roughly 4.5 multiplier on sales (ACTUALLY IT WAS A 5.4:1 RATIO). This is an excellent price for shareholders, especially in today’s climate. The goodwill generated toward Iron Mountain will help ease the trauma of a merger.
• We presume the VCs will get their money back, and the founders will get a small amount of money, but not enough to retire. Everyone else is probably wiped out.
• The $112M was probably largely determined so that VCs and bank-like creditors get their cash back.

All further input welcome: Either post here or email david.ferris@ferris.com. If you want anonymity, let me know and we won’t disclose your identity.

… David Ferris

Anti-spam vendor Cloudmark announced it has acquired Bizanga, which provides messaging software for service providers.

Comments:

• Helps Cloudmark offer a fuller messaging solution to service providers. That makes sense.
• Bizanga executed well in a very difficult, highly competitive market.
• It succeeded in getting some major, high-visibility clients, including Cox Communications.
• Bizanga investors had extremely high revenue expectations that were probably always unrealistic.
• Bizanga had taken about $12M in three rounds between 2003 and 2009. The company achieved a lot with that–most others would have eaten through much more.
• Purchase value guesstimated at $5M to $20M. Bizanga revenues estimated at perhaps $3M or $4M annually; Bizanga never really got into profitability.
• Very unlikely that investors got a decent return; at best they’ll have gotten their money back.
• Bizanga was a delightfully classy operation, reminding us that Man does not live by bread alone. Its gourmet dinners are fondly remembered by anyone lucky enough to have been invited (including your current interlocutor); its customers-and-us souvenir photo albums were also strikingly elegant.

Does anyone have hard information on the deal cut, and Bizanga’s P&L prior to the transaction? Please post here or email david.ferris@ferris.com if you can help–anonymity obviously respected if you want anonymity. Thanks!

… David Ferris

The Experts Conference has a very deep technical lineup of Exchange sessions planned. The event is April 25-28 in Los Angeles. I will be conference chair again this year.

The agenda includes:

• An opportunity to sound off on what you like and don’t like about Exchange Server, and where you think Microsoft is missing out compared with other messaging platforms like Gmail, Lotus Live, Cisco and others. Your feedback will passed on to Microsoft’s Exchange team after the event.
• Depending on Microsoft development timeframes, we hope to be able to run a session on Exchange 2010 SP1.

Other sessions on the agenda include the following Microsoft speakers:

• Scott Schnoll and Ross Smith IV speaking about the many facets of Exchange 2010
• Konstantin Ryvkin speaking about Microsoft’s internal deployment of Exchange 2010
• Brett Shirley speaking about the internals of the Extensible Storage Engine (ESE), B-tree splits, etc.

Unified communications content will include:

• Lee Mackey speaking about his hands-on experiences on many Office Communication Server deployments
• Anthony Vitnell speaking about the challenges of Microsoft/Cisco unified communications coexistence

This will not be a marketing event, but rather a deep technical conference. If you’re planning to attend, send us an email — we’d love to see you there.

David Sengupta. In addition to his role as Ferris analyst, David is Chief Architect for Quest Software, and has been a Microsoft Exchange MVP since 1998.

Exchange 2010 includes archiving and e-discovery; these need powerful indexing capabilities.

Third-party on-premise archiving vendors struggle constantly with indexing technologies. Indices get corrupt and take days or weeks to regenerate. Searches return results that aren’t as expected, or aren’t understood. Indexing technologies age, and when a vendor replaces them, your corporate memory looks very different.

To understand the challenges, think of Outlook:

• It’s often hard to find email in PSTs.
• You frequently don’t get what you’re looking for.
• You get “indexing is not complete” messages.

Now consider e-discovery on a corporate scale. Searches become critically important. For example, you may need to defend your CEO against accusations that might land him in jail; your CEO is certain that an email is there, but the search tool can’t locate it. In the meantime, you have five days to find the email, and the clock is ticking.

Exchange 2010 includes a new discovery module that searches primary and archive mailboxes, and works across multiple mailboxes. It is built by one of the smartest teams in the Exchange product group. However, it’s unclear whether or not the search will be good enough. If it’s like our experiences with Outlook search, the answer is no.

We think Exchange search should be a lot better than that of Outlook. Nevertheless, the challenges are substantial, and there is a good possibility that it won’t be up to the job. For example:

• Important file types may not be supported.
• Documentation may be unclear on how to adjust the index, and when adjustments need to take place.
• Users may not understand the results of a search.
• There may be problems with non-English searches.
• Wildcards and stemming support may be limited.

We would welcome input from readers on their practical experience with Exchange 2010 searches in the stressed and demanding environment of e-discovery.

… David Sengupta

The single most important technology change of our time is happening around the cloud. Entire industries are being changed or made obsolete. The world as we know it is changing fast.

Examples of what has changed:

• Paper-based maps and manual navigation have been replaced with cloud-based maps and GPS systems.
• Photography has shifted from film-based to digital, with processing and often photo albums primarily cloud-based.
• Books are moving from paper-based to electronic paper or digital, with cloud-based ordering and digital library functions.
• Email systems are transitioning from on-premises to cloud-based solutions.
• Print media and advertising have largely gone online.
• Customer relationship management (CRM) systems have moved online.

And examples of what is likely:

• Enterprise telephony will shift from on-premises PBX to cloud-based VoIP solutions.
• Home-based landlines will be replaced by mobile phones.
• Backup and recovery systems will move to the cloud.
• Systems management will move to the cloud.
• Archiving and compliance will move to the cloud.
• E-discovery solutions will move to the cloud.
• Systems with heavy processing requirements will shift to elastic compute technologies.
• File storage will move from local computer-based or external hard drive-based to primarily cloud-based.
• What remains of fax will disappear.
• Human-operator-based conference call bridges will go away.
• Translation services will be mainly automated and go online.
• Paper-based billing will be completely replaced by online.

In each case, the primary technologies and delivery mechanisms will move to the cloud. On-premises solutions will become the exception, not the norm.

As you navigate town with your GPS, BlackBerry by your side, with your new Kindle reading your ebooks to you via text-to-speech as you drive, we suggest you reflect on this. The shift towards the cloud is compelling, and in some ways, irresistible.

…David Sengupta

We’re hearing a series of rumors that something is going badly wrong at Dell/MessageOne. Eg:

• They’ve lost a huge amount of customer archived email over the past couple of weeks
• Many customers are making inquiries about other vendors and their ability to ingest/absorb their historic archive data
• One vendor told us they had been asked to help customers move their emails back from Dell/MessageOne and the most efficient way to ingest large amounts of data (10 TB for example)

Can anyone helps further? If you have input, please send to me at david.ferris@ferris.com or call me on +1 415 367 3436. If you prefer to be anonymous, let me know and we won’t publish anything that identifies you.

… David Ferris

There are three big players in the professional networking business: LinkedIn, Viadeo, and Xing. LinkedIn is best known in the United States. But France’s Viadeo and Germany’s Xing are important competitors. Viadeo recently paid us a visit. The company’s focus is on Europe and emerging economies. It has 25 million subscribers. It plans to enter the U.S. market through leveraging its international subscribers, such as those in Canada and Mexico.

Main services from the user standpoint:

• User directory.
• Users can belong to many different interest groups.
• Each interest group provides:
  • Common bulletin board/wall
  • Directory of interesting events
  • Q&A/forums for advice sharing
  • User-contributed news

Finances:

• Most use is free.
• Around 10% pay for services. This is pretty good compared with LinkedIn, which appears to have around 1% who pay.
• Fee is around 6 euros ($9) per month if you’re a paying subscriber.
• Ferris Research estimates revenues at $40 million annually. The company is profitable and growing at a healthy clip. It’s taken a total of 15 million euros ($23 million) external funding.
• Revenue breakdown:
  • User subscriptions: about 50%
  • Recruiting tools sold to HR and consultants: about 25%
  • Advertising: about 25%

Observations:

• The collaboration tools are conventional but attractive.
• The rich collaborative environment means that in principle, Viadeo can offer extremely targeted advertising, for which advertisers should be prepared to pay handsomely.
• Viadeo’s collaboration tools are an important point of differentation from LinkedIn, which is directory-centric.
• Viadeo’s localizations–e.g., it is currently offered in Dutch, English, French, Italian, Portuguese, and Spanish–are another important differentiator.
• Around 10% of subscribers pay up. This is pretty good compared with LinkedIn, which appears to have around 1% who pay.

… David Ferris

If you care about compliance, the “Drafts” folder in your Inbox needs to be archived.

Archiving vendors take several approaches to email archiving. Some access mailboxes via MAPI and pull items out into the archive. Others intercept SMTP traffic and journal a copy into the archive. Others copy the database transaction logfiles (’log shipping’) and rebuild the email database for archive reasons.

Journaling only captures what has been sent. MAPI and log shipping capture the Drafts folder.

This story on CNN contains a nugget about why archiving Drafts is important. “A Yahoo! e-mail account was set up so the men and militants could communicate …. E-mails were never sent from the account, but people would leave messages in the draft folder and delete them after reading.”

Whenever a law or compliance regime exists, people will try and get around it. Archiving Drafts is necessary for compliance.

… David Sengupta

Exchange 2010’s database strategy is very interesting. The new Database Availability Groups and the benefits they offer for data protection and quick recovery are striking.

Overall, Microsoft is optimistic that mailboxes will be able to grow to 10GB or more. However, it’s unclear how large mailboxes will perform in practice. Several concerns spring to mind:

• By leveraging low cost disk, it is feasible, according to Microsoft, to store email online for up to 10 years. If you assume a typical user stores 5 MB of new email daily, this translates to 1.8 GB of email per year. Allowing for the continued growth in average message size, this translates to perhaps 10GB mailbox sizes over three years. For 100 users, this means 1 TB of total storage for Exchange. For 1,000 users, 10 TB of email storage! Managing multiple terabytes of storage is no simple matter when you take into account disk failures, power consumption, disaster protection, daily maintenance and monitoring, and so on
• Database maintenance and ESEUTIL need space. Offline defragmentation requires 110% free disk space. Thus the amount of storage required should be doubled for maintenance purposes. True, Microsoft advises that offline defragging is unnecessary and inadvisable; hopefully this turns out to be wise counsel
• If we wish to use the new Database Availability Groups for data protection and failover, extra copies will be required. Two to three copies seems likely, which implies another 2-3X in email storage
• Single instance storage is no longer supported, which means message stores can’t be shrunk by 20% or 30% or so

In a typical scenario, for every 1 TB of primary database storage, perhaps another 3-4 TB of disk space may be necessary for various support capabilities, such as Database Availability Groups. This can quickly translate into a large amount of storage.

All in all, I doubt Microsoft’s claims that imply storing TB’s of email data online is a simple proposition.

… Bob Spurzem

Charmingly named YippieMove helps you migrate email from one message store to another. It’s mainly aimed at small businesses and consumers, who are using a cloud-based service. Easy to use, $14.95/mailbox.

It’s just IMAP-IMAP email, no address book or calendar migration services. No, it doesn’t even migrate Notes apps.

… David Ferris