back to article New tech revolution: Small biz begins to lock out industry giants

Software as a Service (SaaS) combined with Virtual Desktop Infrastructure (VDI)-style technologies promise to free us from vendor lock-in once and for all. The consumerisation of IT, Bring Your Own Device (BYOD) and hybrid cloud-based applications are additional marketing buzzwords relevant to this discussion. Each concept is ( …

COMMENTS

This topic is closed for new posts.
Silver badge
Linux

I like the idea

But I'm not convinced.

SMEs are willing to put up with outages and do without features but they are also ignorant to the possibilities and the risks.

However, an outage for an SME is evaluated on the basis of "customers not happy" not "(workers x downtime)" which makes them far more flexible. I worked for a very up-and-coming MS consultancy and their sharepoint systems were forever going down and they swallowed anything a vendor told them. I don't know if the costs & benefits cancelled each other out.

The big suppliers have a problem with scale - its very hard and expensive to do, so a SaaS provider has incredible complexity and uptime requirements where as the local mechanic has an access db which gets copied to a USB stick every day in case the invoice computer goes belly-up. And if the invoice computer does go belly-up, he'll write out the invoices by hand and carry on until his nephew can get down to tesco's and buy a new laptop for him. It is only when management costs start to rise that increasing the "cleverness" and complexity of the management becomes viable. The simplicity of the mechanic's requirements means that his systems have great longevity which means costs can be amortised better than high-requirement environments. It means he can't gain much by going for FLOSS, custom software or more robust systems vs Windows running on Acer, but it also means he has less to lose - he's more agnostic.

I suspect the reason SME's are ignored, is that they are more price sensitive and there are fewer seat licenses to be had. Big complex environments need large numbers of customer licenses to support them, not because the software is expensive, but because the architecture and fixed hardware costs required to scale are expensive.

0
0

Re: I like the idea

That is a little dismissive of SMEs.

Many new and up coming companies, including mine, are tech start ups and the idea of handing over a bunch of cash to Oracle or Sage or even MS is ridiculous as it would eat up any profits we make.

Using Saas for our accounting, BYOD for much of our mobile kit and FLOSS for everything else (with appropriate redundancy) means we are flexible and have low over heads.

We are also as up to date in which tech we are using as any large company.

3
0
Bronze badge

Mainframe/PC deja vu

It's deja vu all over again - I can see exactly the same arguments a few decades back, that nobody would advocate replacing the million dollar three-phase monster with the six figure IBM support contract with mere 'personal computers', it would be career suicide ... meanwhile, the bean-counters were discovering they could get the answers they needed faster on a PC in 123 than from the mainframe. Suddenly, somebody notices the mainframe's not actually being used any more: the PCs have taken over.

At work, we had a six figure SAN with a team of acolytes tending to its every whim, sacrificing goats as ordered in an attempt to keep the clunky 80s email system grunting along. Chucked now, even Microsoft can do a far better job of email than that fiasco, for a tiny fraction of the cost.

Yes, a giant SaaS provider needs to address ... actually, exactly the same problems any other very large outfit would do, which makes it ironic they are capturing the smaller users first.

To a small business, it's often a choice between an SaaS provider's clustered, replicated servers or an elderly Dell out of warranty. Of course the SaaS provider can be much, much more reliable! In a big company, they probably run to multiple servers, some load balancing or failover, decent backups - but 24 hour staffing? We don't, with nine figure turnover and an IT dept headcount well in double figures: it would cost too much. End result is that almost any commercial service is likely to deliver lower downtime, just because problems have to wait until 9-5 M-F to be fixed.

0
0

SaaS just shifts the single point of failure

from the server in the corner to your internet connection.

How many SMEs are running on a broadband product with no SLA, guaranteed uptime or failover option? The big box in the cloud may be well protected, but it a BT engineer pulls the wrong card from the cabinet or the JCB digging up a water main goes through the fibre you're left twiddling your thumbs until someone can be bothered to fix it or else dragging the office to the nearest MacDonalds and hoping the free wifi is still up.

0
0
Gold badge

Re: SaaS just shifts the single point of failure

For $deity's sake man, you're commenting on El Reg. You know enough to come up alternatives! Most people have two broadband providers. Get both. Use a pfsense firewall and do failover + load balancing. Push a button an launch an instance of some IaaS widget that serves as the other point of a VPN tunnel. Then pass your network traffic through your mobile connection.

Send your staff home and have them use their home connections for a day. The stuff’s all in the cloud, why are these people even in the office to begin with?

Unless you are in rural wheresville, options exist. Good enough ones at least to handle the occasional outage brought about by some lummox with a backhoe on an emergency basis. 24/7 is nice-to-have for SMEs, but rarely an absolute requirement.

0
0

Re: SaaS just shifts the single point of failure

So what use are two broadband connections? The failure is almost never at the ISP level, so if you've 2 connections coming along the same trunk from the same cabinet they're more than likely going to be affected by the same fault. To do it properly you'd need two connections from different suppliers coming into your premises using two different routes. All of a sudden what you're paying to make sure you can access your data in the cloud would finance a pretty decent rack of servers in your own office.

SaaS has many benefits, but you also have to factor in the levels of complexity you're adding to your environment. Most may not be obvious and are easy to ignore, but there's a lot of hops between you and your data and with both Google and Microsoft experiencing prolonged outages on their office productivity platforms you'd probably have a more reliable system if you chose to run it yourself.

1
0
Facepalm

Re: SaaS just shifts the single point of failure

"To do it properly you'd need two connections from different suppliers coming into your premises using two different routes."

Exactly. As if your average SME is going to shell out for that. It's hard enough to get the buggers to do backups.

1
0
Gold badge
Pint

Re: SaaS just shifts the single point of failure

@Alan Bourke: The cost of broadband (especially a "Basic" package for emergency use only) is negligible. Yes, even to SMEs. Even to startups that are just a few months old and in the process of acquiring their first customers (as my personal company). It is one cost I have never had a problem convincing any one of my customers about. An additional $40 a month is – all things considered – fairly minor.

As for the bloke spreading FUD about the “unreliability” of cloud passed services…put in your teeth grandpa, we ain’t on yer lawn.

First off: you only achieve high uptimes if you constantly recycle the equipment. Eventually the stuff just gets old. Disks die. RAM failures increase. Even RAID cards start to go. (I should know; my largest customer is sitting on a hardware estate that ranges between 4 and 10 years old, averaging 6 years old.)

Do I have a better uptime than Amazon? Yes. But only just barely; even with active-passive backup systems, (let’s be realistic: few if any SMEs have true HA,) switchovers take time. You have to bring the tech (me) in from wherever they are at the moment, the switchover process has to be completed, and the data verified good before you fire up.

If the particular system you happen to be restoring to service isn’t backed up by an synchronised partner of some variety then the restore process is going to take even longer. You’re in even more poo when it comes to desktop restores. People do save files where they shouldn’t, and SME’s don’t have the money for the software that finds the buggers and moves them. (They also don’t tend to let you lock the systems down far enough that you prevent the users from writing to the local file system.)

Being an SME admin is a complicated job. Business owners and CxO types are often MUCH closer to the decision making and accounting of projects than they are in larger organisations. There are interesting compromises and IT issues that vary per company and you really can’t wrap up “all” SMEs in any nice generalization. You can’t say “well you should do this instead” and wave dismissively nor can you can’t simply treat all SMEs as though their infrastructure requirements, budgets or IT demands were the same.

Well, actually, you can, but you’ll look like an idiot.

A few hours of outage is not the end of the world to most SMEs. Even in the middle of the day, even at the height of busy season. There are workarounds. The critical systems that need computers in-house (say to run the printers or the widget stamping machine or the hullablooo creation mechanisms) will probably never be cloud-based. They will survive a cloud outage.

The point-of-sale systems can be worked around with ten cent pencil, and entering the data when the cloud comes back up. I’ve seen this failover mechanism in use with my own mark one eyeball and it works just fine. Office packages loss for a few hours can be overcome by simply doing something else that is on your enormous pile of shit to do that day and coming back to the office package later.

If the middleware is down, life sucks, but again; ten cent pencil to the rescue! Jotting down the information that staff would normally enter into whatever portion of the middleware they normally use allows them to enter that info when it comes back up. If you are going into graph-and-chart withdrawal from the BI side of the thing being offline for a few hours, get checked out by a psychiatrist right away.

Yes, there is alwys the possibility something customer facing goes down. That would suck. I solve that with a script that took me 5 hours to write that interrogates the existence of critical systems. If it detects an outage, it posts a “we’re sorry, Velociraptors and internet forum commenters ate our servers. They’ll be back in no time. Click here if you want to receive an e-mail when everything’s back online.” I can usually name who is out, and why…customers love it. No complaints.

So I’m right back to “what the hell are you rambling on about?” Judicious use of cloudy whatsit widgetry is a boon to SMEs. Full stop. Quit fighting the future. You’re starting to sound like one of those nutters from ten years ago screaming “virtualisation is a stupid plan. There’s overhead! It’ll never catch on!”

*patpat* Have fun with that.

0
0

what the hell am I rambling on about?

Bitter experience I'm afraid. Our client base runs from 200 user SMEs to 25000 user multinationals with their own Microsoft account managers, with a few systems that absolutely, positively have to be available otherwise people will die, so I've seen both sides of the coin. Our job isn't to flog servers or software, it's to make sure the users can do their work when they need to. If that means a cloud based system is the right solution we go with it, but too often "the cloud" is sold as some panacea to all a customer's ills without properly analysing the full impact of handing over core systems to someone else. I've worked on both good and bad cloud projects and have a few real nightmares under my belt. An ISP screwing up a connection and taking the customer off line for nearly 2 weeks. A lorry crashing into a telecoms cabinet in the street round the corner that took a week to fix. A cloud application where the hosting company woefully underspecced their kit while overselling their service leading to months of terrible performance. And one glorious case where a client fell out with a vendor and it took the threat of court action to get some of their data back so they could move to a new application.

So judicious use of cloudy whatsit widgetry can be a boon to SMEs, but us techies have to be a bit more cynical when it comes to evaluating the options and leave the cheerleading to the sales people.

1
0
Gold badge
Mushroom

Re: what the hell am I rambling on about?

None of the issues you are talking about are exclusive to cloudy apps. A car can take out the power pole at my company and I'm just as hooped. (Indeed, this has happened.) I have had the internet cut for weeks on end, and every single time managed to get an alternative in place within 24 hours. I'm a sysadmin. That's my job.

If you suffer from such binary thinking that you cannot read an article about cloudy apps without believing that they are being touted by the author as a magical cure for all ills then you don't belong on the internet. Or reading newspapers. Or books. Or technical manuals. Or anything, really. If the world is black or white, with nothing in between you really shouldn't be doing much of anything at all, because you are a danger to every single individual, business and most animals you encounter.

On site IT has drawbacks. Cloudy IT has drawbacks. Hybrid IT has drawbacks. No IT at all has drawbacks. Every single thing that we can possibly introduce to solve any problem – from keeping predators away at night to making picosecond financial transactions – is a series of tradeoffs. Stability, reliability, redundancy, capex, opex, skills availability, vendor lock-in and yet more.

There is no magic bullet. There are no perfect solutions. But when someone highlights the benefits of one particular solution without a doctorate-level comparative treatise on the possible drawbacks and comparative analysis for every conceivable use case they are most emphatically not advocating that particular solution as the magic solution to all ills.

In the case of this article, I can assure you that your less-than-humble scribe made some assumptions about the general level of intelligence, competence, experience, knowledge and comparative analytical abilities of the intended and likely audience.

If you have failed to meet my (apparently too lofty) expectations, then I deeply apologise for the incursion into your worldview. You have enlightened me; I shall promptly redouble my efforts to target my articles squarely at utter mediocrity.

0
1

Re: what the hell am I rambling on about?

Now now Trevor, no need to be so defensive. I never claimed non-cloudy computing was the only answer. It's just I've been in the trenches too long to buy into another revolution so easily, and while you can brush of slower, feature-limited applications as something SMEs are willing to accept I've been the mug dragged in front of the chief executive to explain precisely why their new super duper cloud application isn't as fast and can't do as much as the local application they used to run in-house.

The thing is, this stuff isn't new. We've been running hosted apps for our customers for years, back before some bright spark came up with "The Cloud" as a cool way to market it. So yes, hosted services are a good option for SMEs, but if you choose to ignore some very real problems with the technology you shouldn't get so worked up when those of us actually rolling out these systems think it's worth mentioning that the picture might not be as rosy as the one you paint.

1
0
Gold badge
Facepalm

Re: what the hell am I rambling on about?

I'm not defensive. I'm exhausted. And the issues you raise - along with many, many more - have been discussed at length in my other articles, comments and so forth. If you want to get snarky that I am not pandering to your pet prejudice, I'll point you in the general direction of an Apple article.

I am not opposed to constructive criticism at all. I am however getting fairly sick of sour grapes. The article was targeted at IT professionals. You know…The Register’s audience. It isn’t FOX news’ technology section, now sponsored by Best Buy and featuring a free lobotomy to bring your IQ below 100!

That any technology – cloud or otherwise – has downsides does not need to be spelled out explicitly in each and every article. You are expected to be smart enough to know that.

Indeed, what I am presenting is “the world as I see it.” (Blog, eh?) This is what I am seeing on the ground. How people are using technology. The things they worry about. And the gist of the article – in case you missed it – is that they are a lot more worried about vendor lock-in and getting the shaft regarding licensing than they are the occasional hiccoughs inherent in a cloud delivery model.

No tech is perfect, but these are the choices I am seeing real people in the real world making. They give me ~500 words. (With some fudge factor when I go over.) I chose to report what I see rather than take up a bunch of that rehashing the same “cloud vs. local” argument again for the 10,000th time.

I’m not going to waste my 500 words a week revisiting issues I feel have already been beaten to death umpteen times, and that my readership could argue from multiple angles in their sleep. They deserve better.

1
1
This topic is closed for new posts.

Forums