Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

PostgreSQL vs. MS SQL

pg-versus-ms.com

Post reply on HN
Print view View original

41–50 of 106 posts

Re: PostgreSQL vs. MS SQL

#41

Post by keithwarren »

general_failure wrote: post #27

I scrolled to a random page. It was totally absurd, stopped right there. "Crucially, because open-source software tends to be written by people who care deeply about its quality (often because they have a direct personal stake in ensuring that the software works as well as possible), it is often of the very highest standard (PostgreSQL, Linux, MySQL, XBMC, Hadoop, Android, VLC, Neo4JS, Redis, 7Zip, FreeBSD, golang, P…

I know numerous guys on the SQL Server team and it actually causes me a bit of anger when I read a clueless statement like that, for the author I feel sorry for his ignorance; the anger comes from the fact that lots of clueless people will actually read this and believe it to be authoritarian because of the verbosity of it alone.

Re: PostgreSQL vs. MS SQL

#42

Post by keithwarren »

UnoriginalGuy wrote: post #26
keithwarren wrote: post #21

Earlier quoted context omitted.

see rest of comments, it is not taking long for people to point out the flaws in the OP

So that is a "no" then? You make a blatant argument from authority and then when asked to affirm your opinion with facts rather than claimed qualifications you refuse (saying "read the other posts" is a refusal if I've ever seen one).

In all honesty it would take hours to write a proper rebuttal. There are many factual errors but there are also many errors that are mostly incorrect. CSV import/export is one of the first things he calls out and he never mentions BCP or SSIS, which are the methods for CSV import in SQL Server. It would be like saying Apache does not support SSL - it is just factually wrong.

Re: PostgreSQL vs. MS SQL

#43

Post by sergiosgc »

blahblahOOO wrote: post #10

As a DBA, that has years of experience with both. I disagree with you. I stopped reading after the very first bullet, and just skimmed through most of the titles. First, you can import/export csv's. Hell all you need to do to export, is right click the view and hit export, then select .csv. Second, you don't want such a simple installation, MSSQL has several options/packages/configurations. If you don't need it you t…

> First, you can import/export csv's. Hell all you need to do to export, is right click the view and hit export, then select .csv.

Are you saying there is no programatic interface for exporting csv? You must rely on the administration tool to generate the csv for you???

Re: PostgreSQL vs. MS SQL

#44

Post by jordigh »

threeseed wrote: post #13

The author is pretty misinformed about how enterprises work. a) Security is critical. It routinely trumps common sense and evidence. And telling people "not to fall for it" is advice that will get you nowhere. b) Vendor support is critical. And no your local mom+pop consulting shop doesn't count. PostgreSQL could really do with solid, global companies like Microsoft, Datastax, Mongo etc who offer training and proper…

What vendor support do you get exactly from Microsoft? Their EULA typically has a SHOUTY CAPS NO WARRANTY CLAUSE, so what does the SLA cover? Have you successfully resorted to the SLA to handle problems you've had?

I'm honestly curious. I just don't know how SLAs work, and whom can you blame if you have problems with MS SQL.

Re: PostgreSQL vs. MS SQL

#45

Post by threeseed »

x0x0 wrote: post #35
UnoriginalGuy wrote: post #31

Earlier quoted context omitted.

While I kind of agree, what exactly is the "right tool" for storing large blobs? You can use the filesystem but then you have to manage data migration/sync by "hand" which is kind of a bitch for distributed systems. Honestly storing large blobs with meta-data telling you about the blobs is extremely common. I'm yet to hear of how you're meant to do it correctly.

commonly, s3 or zfs nas for storage and pg for metadata

Commonly ? No not even close. The most common solution is that people just dump it as a BLOB in whatever database they are currently using. Which based on numbers is likely to be MySQL.

Re: PostgreSQL vs. MS SQL

#46

Post by sergiosgc »

UnoriginalGuy wrote: post #31
DrJokepu wrote: post #18

Earlier quoted context omitted.

That being said, if your relational database has 1GB records, you really need to reevaluate whether you are using the right tool for the job.

While I kind of agree, what exactly is the "right tool" for storing large blobs? You can use the filesystem but then you have to manage data migration/sync by "hand" which is kind of a bitch for distributed systems. Honestly storing large blobs with meta-data telling you about the blobs is extremely common. I'm yet to hear of how you're meant to do it correctly.

On pgsql, the blob type. It stores the object out of the row store, but still in the database.

Re: PostgreSQL vs. MS SQL

#47

Post by blahblahOOO »

sergiosgc wrote: post #43
blahblahOOO wrote: post #10

As a DBA, that has years of experience with both. I disagree with you. I stopped reading after the very first bullet, and just skimmed through most of the titles. First, you can import/export csv's. Hell all you need to do to export, is right click the view and hit export, then select .csv. Second, you don't want such a simple installation, MSSQL has several options/packages/configurations. If you don't need it you t…

> First, you can import/export csv's. Hell all you need to do to export, is right click the view and hit export, then select .csv. Are you saying there is no programatic interface for exporting csv? You must rely on the administration tool to generate the csv for you???

@sergiosgc Nope. There are several ways of doing this. Using the GUI is the easiest, and the most common method. You can use SSIS, or the bcp command.

For example:

bcp "SELECT blah... FROM blah" queryout C:\..\blah.csv -c -t, -T –S

Re: PostgreSQL vs. MS SQL

#48

Post by Spearchucker »

jordigh wrote: post #44
threeseed wrote: post #13

The author is pretty misinformed about how enterprises work. a) Security is critical. It routinely trumps common sense and evidence. And telling people "not to fall for it" is advice that will get you nowhere. b) Vendor support is critical. And no your local mom+pop consulting shop doesn't count. PostgreSQL could really do with solid, global companies like Microsoft, Datastax, Mongo etc who offer training and proper…

What vendor support do you get exactly from Microsoft? Their EULA typically has a SHOUTY CAPS NO WARRANTY CLAUSE, so what does the SLA cover? Have you successfully resorted to the SLA to handle problems you've had? I'm honestly curious. I just don't know how SLAs work, and whom can you blame if you have problems with MS SQL.

Vendor support from Microsoft is about requesting help with issues. You get x number of support calls, depending on what you paid. Microsoft has fixed service levels for that, and you've little option other than paying more money if you want a quicker turn-around.

SQL Server itself doesn't ship with an SLA.

You'd get that from your service provider or systems integrator, because the customer chooses her own service levels (determined by business continuity objectives). Her SP or SI (or in-house IT department) would then deploy SQL Server in a way intended to meet the required service level. Specifically, the customer might opt for five nines (99.999%) availability. Such a deployment doesn't look anything like one that must only be available from 8am to 5pm on weekdays - even though both SLAs are met using the same SQL Server code base from Microsoft (but on vastly different hardware and network configurations).

The availability associated with an SLA usually goes hand-in-hand with disaster recovery (recovery time objective, and recovery point objective[1]), but can also apply to support turn-around (a support request is triaged, and based on severity is resolved within the amount of time specified for that severity by the SLA). As mentioned, if I'm not mistaken Microsoft has fixed service levels for that. IIRC an MSDN subscription gets you a small number of free requests. An Enterprise Agreement gets you a whole bunch more.

There are many other quality objectives you can specify with an SLA, including efficiency (capacity), integrity (security), and robustness (stability) [2].

So unless you're hosted by Azure or have an Enterprise Agreement, Microsoft is rather unlikely to provide you with an SLA. And even then you'd be the one telling them what your service level requirement is. If Microsoft can't deliver on your service level requirements, you'd do it yourself, or get a systems integrator to do it for you.

[1] http://www.druva.com/blog/understanding-rpo-and-rto/

[2] https://www.wittenburg.co.uk/Entry.aspx?id=d8c54975-bd0a-410...

Re: PostgreSQL vs. MS SQL

#49

Post by SideburnsOfDoom »

threeseed wrote: post #39
UnoriginalGuy wrote: post #31

Earlier quoted context omitted.

While I kind of agree, what exactly is the "right tool" for storing large blobs? You can use the filesystem but then you have to manage data migration/sync by "hand" which is kind of a bitch for distributed systems. Honestly storing large blobs with meta-data telling you about the blobs is extremely common. I'm yet to hear of how you're meant to do it correctly.

I had to solve this problem at my last job and we had vendors let us know how companies like Apple, Sony, Disney, EA etc have solved it. Basically there are two ways to do it. (1) Store it on some "filesystem". In quotes because how you do it can vary wildly e.g. S3, GlusterFS, Standard directories using DRBD for HA. (2) Take the blob, slice it into pieces, hash it and spread it across a sharded database. Generally i…

We found that a GlusterFS filesystem is a good way to go for our case of 100s of terabytes of unique files with some failover. Amazon S3 would simply cost far more. The file metadata is in the RDBMS, obviously. But storing the binary data in the RDBMS wouldn't be a good idea, and would choke long before it got to our scale.

Re: PostgreSQL vs. MS SQL

#50

Post by rbanffy »

keithwarren wrote: post #41
general_failure wrote: post #27

I scrolled to a random page. It was totally absurd, stopped right there. "Crucially, because open-source software tends to be written by people who care deeply about its quality (often because they have a direct personal stake in ensuring that the software works as well as possible), it is often of the very highest standard (PostgreSQL, Linux, MySQL, XBMC, Hadoop, Android, VLC, Neo4JS, Redis, 7Zip, FreeBSD, golang, P…

I know numerous guys on the SQL Server team and it actually causes me a bit of anger when I read a clueless statement like that, for the author I feel sorry for his ignorance; the anger comes from the fact that lots of clueless people will actually read this and believe it to be authoritarian because of the verbosity of it alone.

The SQL Server group (and their product) is one of the very few things in Microsoft that has earned my deepest respect. Not Windows, not Exchange and, certainly, not ShamePoint.

I frequently rank Microsoft's product line as follows:

1 - Natural keyboard - the one there is no reasonable replacement for (the kinesis keyboards are 5 times as expensive). If there is a reason my none of my machines is 100% Microsoft free, this is it.

2 - their mice - simple, precise, comfortable, inexpensive

3 - SQL Server - it almost makes having a Windows server around worth the pain.

Having said that, I have enormous respect for PostgreSQL. It's a very solid RDBMS and it's my database of choice most of the time. It is more comfortable to use from a command line than SQL Server will ever be (which is OK, because a CLI is not a high priority over there).

Post reply on HN