Tag Archives: cloud computing

Migrating to the cloud – Small Business

Posted on by

This is probably a bit out of order since I should cover some foundation information first but I know a few of my readers are most interested in this part of the series so I figured I would start here first.

As a general definition, I classify Small Business as an organisation with less than 100 people who generally don’t have any full-time IT staff.

I’m going to put a Mantra here that I will probably repeat every post in this series:

Cloud Computing is not about saving money! Cloud Computing is about improving business efficiency which will allow you to make more money!

The Cloud, or Cloud Applications are probably of most immediate interest to Small Business. Move everything to the cloud and stop worrying about having to buy a “Server” (whatever that is) every 5 years, or to pay the IT Guy 300 bucks every time he has to fiddle with something! Won’t have to change those damned tapes we pay $100 each for and don’t do anything! Yada Yada Yada.

Unfortunately, what most small businesses think they understand about Cloud Computing is utterly and completely wrong. This is pretty much what the majority of small businesses understand about most aspects of ICT which is one of the reasons they can often be so hard to deal with. I’m going to digress and use an analogy a friend told me years ago. He was a former electrician, he gave it up in frustration about customers complaining to him about home much he charged. They didn’t care that he had to crawl under the house or into roof cavities full of insects or other pests. He had to drill through double brick and then re-fill the holes. All they saw was a power point on the wall that cost them $500 to install. ICT in small business is a lot like that; many small businesses do not understand the time or effort it takes to keep things working so they can continue to do business. To them, cloud computing offers a way to get out of paying for people to look after their infrastructure. To be fair, a lot of marketing material promotes this ideal. The reality in unfortunately far less utopian.

There is one thing every small business needs to understand about Cloud Providers. They don’t care about you! They only care about your money!! I’m sure some idealistic Cloud Service offerings may initially have their customers’ best interests at heart, but if they try to provide the sort of personalised customer service that most small businesses expect to receive, they will quickly find themselves out of business. To make offering Cloud Services to small businesses worthwhile requires that they attract a lot of customers. When you have a lot of customers, the most expensive part of your business stops being equipment and facilities and starts being human resources; particularly customer support. Because of this, you will still need to rely on your IT support people to both look after your internal systems as well as to interact with your Cloud Service Provider.

Small businesses generally only have the option to go S-cloud (see my intro post for definitions). The only way cloud providers can offer them economies of scale is to bundle their infrastructure along with a lot of other clients’. This unfortunately leads to the biggest issue with S-cloud which is lack of flexibility! When you start sharing infrastructure with a bunch of other people, that starts limiting what you can use that infrastructure for, and that’s where things start to come unglued for small business. Here are some of the reasons:

  • There are very few types of business that don’t run some type (or often multiple types) of industry-specific software. It could be inventory management, it could be diagnostics; it could be almost anything. The odds are that you will not be able to efficiently take those applications to The Cloud. This means you will still need a File Server and still need someone to provide support (in fact you will always need someone to provide support). Now The Cloud can only deliver some applications.
  • When you move any part of your business to The Cloud, you will have to take your company Internet link(s) far more seriously. You’ll have to say goodbye to that $50/month ADSL link and invest in some serious bandwidth. If you put in anything less than around 10Mbps (which will generally cost around $500-1000/month in Australia), your user experience is going to plummet. In a localised environment, just about the only “real-time” traffic that uses The Internet is web-browsing and even much of this isn’t really real-time as web browsers can compensate for delay fairly well. Any application that is delivered via The Cloud is completely real-time which means if you don’t have a decent Internet link (I’m keeping things simple here and not going in to private networks for now), the performance of those applications is going to be dreadful. They may be “thin apps” (look it up if you don’t know what that means) but you can still only squeeze a certain amount of toothpaste out of the tube through a tiny nozzle. Now The Cloud means I have to pay more for my Internet link.
  • Probably one of the biggest attractions of The Cloud to small business is Voice. Not having to replace an expensive PABX, and being able to consolidate all their branch offices into a centralised number range and not pay for calls between them in return for a “reasonable” ongoing fee can grab attention. Unfortunately, Voice is currently probably the worst application generally delivered through The Cloud (though this is starting to improve). Most “Cloud Voice” solutions are delivered through a mechanism called “tenanting” where a centralised PABX is carved up to be delivered to multiple clients. This technology was originally developed for Serviced/Virtual offices and is very good for that. Unfortunately, a tenanted systems lacks a few key features that can be disastrous to any business. Possibly the most insidious issue is the losing the ability to “night switch” the PABX. For those who don’t understand telephony, a PABX generally has 3 modes: Day, Night1 and Night2 and they are critical to most organisations. Night Switching a PABX (which can be automatic or manual) is the mechanism used to globally change the behaviour of the system. Night1 is normally used for the “Thank you for calling XYZ co. Our office is currently closed…”. Night2 is used for special events (e.g. public holidays). Losing this feature is not an option for small business. The other thing you lose with a tenanted system is the ability to deploy a proper Reception Console which – despite what a lot of people thing – is still used for high-volume call intake at the majority of businesses (you also lose call centre capability…). Try giving a receptionist a standard phone and then start running while she looks for her shotgun. There are a lot of other issues prevalent in many “Cloud Voice” system but I’ll stop there. incidentally, when you amortise the cost of Cloud Voice against what you can get a local system for, you generally end up paying much more (gotta love those 3-5 year contracts!). Now The Cloud gives me expensive, crappy phones.

Those are some of the issues with small business moving to The Cloud. Now let’s consider some of the positives when it is done right:

Firstly, don’t expect to pay your IT support people any less (and IT support people, don’t expect to be out of a job). They will be just as necessary (if not more so) to keep things going.

The right solution for small business (and indeed for most businesses) is a Hybrid-Cloud environment. Many applications can move to The Cloud but others need to stay behind. The following should always be in The Cloud:

  • General business applications (MS-Office, most major accounting software, etc.)
  • Data storage (more on this in a bit) including backup
  • Remote Access/VPN
  • Email
  • Intranet

The following applications should never be in the cloud:

  • Any application that requires local, direct input/output (i.e. from a scanner, plant equipment, etc.)
  • Desktop Antivirus (although you must make sure your Cloud provider includes antivirus protection a well)
  • Any heavy graphic applications (CAD, Desktop Publishing, video production, etc.)
  • Any customised business application

Some of the above is probably fairly obvious but lists can be helpful when deciding where to put what. I’ll cover a few of the above points in more detail:

Data Storage is probably the single most important and beneficial aspect of Cloud Computing. An organisation’s data is its lifeblood, and local backups are generally slow, tedious, expensive and often unreliable. I doubt I’ve ever encountered a business over 5 years old that hasn’t encountered some amount of data loss in their past. Cloud Storage is relatively inexpensive and extremely reliable and portable. An example of this (which I’ve seen many businesses use) is something like DropBox (unsolicited plug for an awesome product: www.dropbox.com). Dropbox maintains replicas of your data both locally and in The Cloud and is extremely good at keeping it synchronised. This means your local and remote applications can access your data equally efficiently. It also means that mobile workers can have all or – more often – a subset of their data replicated to their laptop and even allow them to access it on most tablet or smart-phone devices. Properly set up – usually with 3rd party encryption tools for added security – Cloud Storage is virtually impossible to lose. A business owner can even keep a replica at home for virtually no cost if they so choose to be completely certain of its protection. A service like DropBox with 50GB of online storage and unlimited deletion recovery costs about $135 a year. There are business-focussed services that cost only slightly more making this the most inexpensive Cloud application as well as the most important. Now The Cloud can offer me almost 100% data protection

One of the biggest advances in Cloud Technology is the delivery of applications. Microsoft, Citrix, VMWare and others have all made huge strides in making applications available securely via Web Browser, which allows them to be accessed virtually everywhere. Being able to look up stock while on the beach, or edit a document while at the airport (and have it immediately available to colleagues) are effortless benefits of Cloud-based applications. Through Cloud and/or web delivery, user platform becomes far less relevant as the actual application processing happens at the Service Provider end (so you can do an MYOB query or fully edit an Excel Spreadsheet on your I-Pad for example). This offers a tremendous increase in flexibility and mobility without the huge costs that used to be associated with running this technology in-house. Now The Cloud allows me to access my business applications from literally anywhere

Whilst I touched on the negative aspects of Cloud-based telephony earlier on, that was mainly to highlight the many shoddy services still being offered (which many businesses are now stuck in contracts with). Done right, Cloud Voice or – more appropriately – Cloud Unified Communications can offer tremendous benefits to any organisation. Users can be given a single-point of contact that can follow them anywhere (to their mobile, to their home, to their hotel room, etc. – for more information on where this is going see my earlier post: http://michaelgwolff.wordpress.com/2012/02/04/microsoft-lync-get-used-to-it-get-used-to-loving-it-because-if-you-are-in-business-you-will-be-running-it-soon/) for virtually no additional cost. With the “right” cloud offering, users of even very small businesses can be granted access to all the features normally reserved for large companies (e.g. video conferencing and collaboration, teleworking, large conference bridges, etc.) without having to give up the most important features detailed above. In my experience, the best Cloud Voice offerings are built on either Mitel technology or Microsoft Lync/Exchange. Now The Cloud allows me to have every UC feature imaginable.

The Cloud is the right direction for the majority of small businesses. The move to The Cloud needs to be made for the right reasons though, and the right 3rd party advice should be sought regarding the suitability of such a move. Businesses that take advantage of The Cloud will find their environment to be far more secure and flexible; and their users being granted far more flexibility. I’m going to write about to look for in a Cloud provider in a later post but as a general rule if your current IT support company/advisor tells you to not move to The Cloud or to move everything to The Cloud, speak to someone else since they probably don’t know what they are talking about.

Michael

 

Note: This is a very complicated topic. Each post is somewhat relevant to all the others and I am writing them one at a time so expect some revisions to come as I develop and refine things further

 

About these ads

Migrating to the cloud – introduction

Posted on by

Hi all,

Based on the feedback I’ve received, it seems most people would like to know my opinion as to why businesses haven’t taken up cloud services anywhere nearly as quickly as most analysts have predicted. This is despite the fact that most cloud solutions offers big increases in ICT efficiency and flexibility whilst promising a very rapid ROI. In my opinion, there are numerous reasons for this; and the reasons are different depending both on the type of Cloud application or service and the type or size of business.

Over my next few entries, I’m going to examine some of these reasons and maybe help provide some road map assistance for organisations that are struggling to make this transition. I’m going to try and cover off the following broad topics (in no particular order) at the very least:

  • Small Business
  • Large Business
  • Hosted VoIP/Unified Communications
  • Security Concerns
  • Design Constraints
  • Transition issues (probably the most important topic)
  • Migration issues
  • Disaster Mitigation (another important topic)
  • Choosing a cloud provider

If you have other topics you would like me to cover, please let me know and I’ll try my best.

For now, I’ll start with some basic definitions. One of the biggest general constraints of “Cloud Computing” is that few people in the industry can actually agree on what the term actually means. Rather than weigh in on the ongoing debate, I’ll offer the following general definition: Cloud Computing is a state where applications and data have become fully abstracted from the hardware upon which they reside. In other words, a true “cloud” offering must be completely portable with minimal effort, and freely (and securely) accessible from virtually any public network location.

A few other definitions I’ll add (as I will  be using these terms in my following posts):

  • I-Cloud – a cloud environment that exists completed within an organisation’s premises or controlled infrastructure
  • E-Cloud – a cloud environment that exists outside an organisation’s premises or controlled infrastructure but who’s components are dedicated solely to that organisation’s use (managed by a 3rd party)
  • S-cloud – a cloud environment that exists outside an organisation’s premises or controlled infrastructure who’s components are shared with other organisations (managed by a 3rd party)
  • P-cloud – a cloud environment that exists outside an organisation’s premises or controlled infrastructure who’s components are shared with other organisations (managed by multiple 3rd parties)

I hope you find this series interesting and look forward to your comments.

Michael

In with the Old out with the New

Posted on by

I’m sure that I’m not the first person to cover this topic but I thought I’d add my bit.

One thing that amazes me is the cycle of technology that goes on. The old cliche of “put the old dress in the closet and 20 years from now it will be back in fashion” seems to pretty much apply equally with IT. I think it was seeing those “reto-handsets” they are offering everywhere for the IPhones that got me thinking. Let’s take a few examples..

The first “real” laptop computers (from about 1989/90 or so) came out with specs more or less similar to a conventional desktop with screens between 14-15 inches. The in the late 90s, there was a rage of “Micro Notebooks”. I had an IBM x240 which was about 8 or 9 inches wide, and I was given an original Toshiba Libretto (Pentium 166) as a keepsake from a friend a few years back (still the smallest production notebook I’ve even seen and it still works perfectly). They were popular for a while but quickly became stale as people realised they didn’t have the horsepower to do any real work on and had crummy battery life. So we moved to the “ultra portables” – with the Toshiba leading the way again with the Portege – which were slightly bigger, had much more grunt and better battery life, and a much bigger pricetag. Execs and graphic designers had them but the majority went back to notebooks with specs more or less similar to a conventional desktop and screens between 15-15.4 inches.

Along comes 2007/8 and suddenly everyone needs a “Netbook” (i.e. Micro Notebook). They can’t build them fast enough and every vendor jumps on the bandwagon. In Australia, the government gave away literally tens of thousands of them to high-school students as part of their “Digitial Education Revolution” funding program…only to realise that they didn’t have the horsepower to do any real work and had crummy battery life (this was particularly depressing for students as they tend to be power users and finding out that you can’t edit photos or videos on your brand new notebook was not a good thing). This buzz this year is for “ultra portable” which are slightly bigger, have much more grunt and a much bigger pricetag. Sound familiar?

About 15 years ago, Microsoft brought out a product called Windows CE (sometimes horribly abbreviated to WinCE) designed specifically for portable computing devices, often employing touch-screens. At around the same time, Apple brought out their Newton portable touch-screen computing device. There was a brief flurry of interest and activity with many vendors (most notably HP) bringing Windows CE devices to market and many die-hard Apple fans swearing by their Newton. Eventually, Steve Jobs killed off the Newton when he took back the reigns of Apple and set about transforming the company whilst Windows CE is still around but used mostly for devices like Thin Clients. About 18 months ago, Apple released a device called the “IPad”, and once again we are off and running with touch-screen computing devices (AKA Tablets).

Finally (for this article) there’s cloud computing. 25+ years ago pretty much all business computing was centralised. Big Data Centers housed monster computers that consumed so much electricity and produced so much heat that they needed dedicated power distribution and cooling separate from other buildings around them. Everyone connected to them via relatively slow data links, which didn’t matter too much since they did all the work centrally. Then we got the PC which placed a reasonable fraction of that processing power directly on to the business user’s desktop allowing them to work discretely without having to share the central computer with 100s of other people; followed shortly thereafter a slightly more powerful PC that was called a “File Server” which further allowed businesses to localise their data and processing all the way down to the branch and individual user level. The Information Sprawl was thus born and the amount of data and information gathered and collected (and dispersed) started growing exponentially as more and more devices were purchased and deployed until the business reached a state where virtually no organisation could even come close to accounting for where all their data is actually located. Around 1998, a company called VMWare popped up, promoting technology that allowed companies to carve up their existing File Servers into multiple “virtual file servers”. They called this technology a “Hypervisor” (they didn’t invent it, but they were certainly the first company to make it work). At the same time, a company called Citrix was starting to gain decent acceptance of their “Winframe” or “Metframe” product which allowed businesses to carve up their existing File Server into multiple “virtual desktops” that users remotely accessed in order to gain access to company applications and data. Both these technologies developed a reasonable following but were limited to use by very large organisations since even though they didn’t need much network bandwidth to operate (i.e. the capacity of the link sitting between where the user and file servers were located), they needed more than most home users, small or even medium businesses had available (i.e. generally dial-up links). The we finally got the last main piece of the puzzle which was the wide availability of high-speed, low cost data links to almost everywhere in the world. So now businesses are moving all their data and processing into Big Data Centres which has monstrous arrays of computers that comsume so much electricity and produce so much hear that they need dedicated power distribution and cooling separate from the other buildings around them.

What’s next on the technology cycle?