Reprinted from my editorial in Database Weekly.
When I was a junior DBA many years ago (even before the company I work for now, Red Gate Software existed), I ran across some third-party software for SQL Server that I thought could really benefit my productivity and save my organization money. I presented my case, proving how the new software would be beneficial to the company. My boss agreed with that the new software would indeed boost my productivity, he refused my request. When I asked why, he answer was “We are a 100% Microsoft shop, and I don’t want to introduce any third-party software into the mix.” He continued to say, “If we should every have any problem with the third-party software, then Microsoft and the third-party software company would point fingers at each other, and the problem would never get solved.”
I understood that my manager was being extremely conservative and didn’t want to risk any potential data corruption, but I thought he was just a bit too conservative. Time actually proved my point. The particular product soon became a market leader, and I have never heard one time ever where the product ever caused data to be lost.
What reminded me of this old conversation was a recent discussion I overhead where one DBA said that he would never take any risk his data to any third-party product that interacted directly with SQL Server data. He seemed to an incarnation of my old boss. The other person in the conservation said that as long as the product was thoroughly tested, and proved safe, that why shouldn’t it be used.
So here’s my question to you this week. Do you fall into the camp of very conservative DBAs that don’t want any non-Microsoft product to touch their SQL Server data, or do you fall into the camp that if a product is proved safe, that it should be used if it can boost productivity and save your organization money? Share with us your opinion, and tell us why you believe the way you do.
Brad M McGehee 
Well, I am definitely a pro-MS guy, and always seek to leverage existing MS products and features out-of-the-box. But, as others have written, sometimes, there is a real and practical need to go with a 3rd party solution, that is not necessarily Microsoft.
One obvious example is VMWare – the ratio of companies virtualizing and deploying their SQL Server infrastructure on VMWare, is much greater than Hyper-V.
Just my 2-cents.
Rober Pearl
“It Depends.” There are a lot of things Microsoft does pretty decently out of the box, and for the most part… out of the box solutions are good enough. But there are other times where Microsoft completely sucks at something and you have to turn to 3rd party software. For example, Source Control. Look at what Microsoft offers you for SQL Source control vs what Red Gate SQL Source control brings to the table. There are not as many players in that arena so it makes for an easy comparison. On other areas with lots of players, it can be a bit draining to ‘find the best one’ and that is where trusted sources come in to help filter the search.
Horses for courses, I think. It depends a lot on the nature of your company business, how real your need is to explore alternative products, and how much resource you have. I’ve worked for many different companies over the years, including international investment banks, insurance companies, mobile phone operators and small, privately-owned technology companies. I’d say each one would have a different viewpoint on this question.
I’ve worked for small companies where they’ve been only too keen to experiment and explore not only new products from other companies, but have been early adopters of Microsoft products too. Bigger players, such as the banks, tend to be a little more conservative, as they can’t afford to take too many risks with tried and tested procedures and products.
On the other hand, my current employer not only uses 3rd party products (although with a healthy degree of scepticism for their abilities as so many of those products are not actually very good in the way they do what they do – no offence Red-Gate!) but we also hack Microsoft’s own products too (such as SQLServer replication; re-writing internal stored procedures and automating replication fault recovery, for example). This approach could be disastrous if it weren’t for a) all my DBA colleagues being highly skilled and competent, and b) having a very large team of DBAs. If you want to be adventurous, you need to do it properly and have the resource to do so.
I realise this is more of a comment on conservatism than 3rd party products in general (as some of those can make the job simpler and less-risky). I find it especially interesting just how much some companies will stretch products to their absolute limit rather than being content to just use them within their advertised constraints.
Having been a SQL Server developer and DBA for nearly 15 years, I have worked with many different database management systems over the years, not that it makes me any kind of authority on the subject.
However, I have seen and worked with several third party tools. Some are great, right out of the box. Others, not so great. 3rd party tools when needed, are usually the market leaders and are there, in most cases because the product fills a need that Microsoft hasn’t yet filled. As with a lot of products, Microsoft tends to see first how the 3rd party product is accepted by Microsoft’s own audience in the market place. In some cases, for instance, backup compression, Microsoft waited until SQL Server 2008 to introduce their own version, but it was available several versions back by Red Gate Software.
Many, actually most, IT managers do not do any investigation into the products proposed by their subordinates in database administration. They know Microsoft SQL Server is the best database management system, so they shop around and get the best price they can and then buy the licenses from the least expensive vendor. No problem there, it is the best buy for the money. It is a Microsoft product and no one can fault them for sticking with Microsoft can they?
Then a few months down the road, the top DBA quits. So the Manager looks around, hires the next best DBA he can find, but the new person suggests purchasing the SQL Toolbelt from Red Gate and SQL Doctor from Idera. The DBA bases her justification on the fact that the backup compression will save 90% of the current disk space used for backup storage. SQL Prompt will save the other DBA’s a tremendous amount of time, because they don’t have to look up all those functions, table names, etc. All the databases are versions 2000 or 2005, and there are 14 Servers and 54 databases. The new person is the lead DBA and there are 2 junior DBA’s. One of the junior people has just graduated from college with a degree in Business Management and wants to do that rather than any database administration. The other person has 4 years with the current company and no training in Database Administration.
Low and behold, the new person wants to buy a product called SQL Doctor from Idera, so the other DBA’s can learn more quickly and efficiently. Also, she can quickly find problems and then either fix them or assign them to the others on the team.
The manager gives the new person time to present her case for purchasing the two tools, and then tells her he will consult with some outside vendor or consultant. In this case the manager actually does call and speak to a person who he considers to be an “EXPERT”. Turns out the expert is an experienced DBA for another company who is a Microsoft Extremist. In his opinion, nothing that is not Microsoft will work properly on “His Data”. So the manager just waits until the new person asks about the proposal. He then says he has consulted with an EXPERT who advised him not to use Non-Microsoft tools on their data.
Even when the manager is shown that the 2 proposed 3rd party tools will save the company over $100,000 per year by not having to hire more people, he says NO!
So, I say you should check out how a 3rd party tools works, test it thoroughly and if you can cost justify it, IT management should attempt to purchase the tool for you.
The manager’s last statement “Well you know Microsoft will have a better solution in their next version”. How long will that be, 2 years? 5 years?