A Fine Slice Of SQL Server
Steve Jones walks through a Python tutorial:
However, there were other errors, which I suspect are related to Python 2.7 v Pyhton 3.5. Rather than solve those, I went on to the columnstore demo. In this, you create a table with 5mm rows and then run a query against it from Python. I did that, then created the columnstore index, then ran it again. The results are below.
And within an hour or so of starting, Steve has hit the 2.x vs 3.x mess in Python.
Comments closedJames Serra links to the Azure Total Cost of Ownership calculator:
For a long time clients would ask me how to determine the cost savings by migrating their applications and databases to Azure. I never had a good answer until now: The Total Cost of Ownership (TCO) Calculator. Now in preview, just provide a brief description of your on-premises environment to get an instant estimate of the cost savings you can realize by migrating your application workloads to Microsoft Azure
I think this is a useful start, though a big part of the value I see in Azure is moving certain things from IaaS to PaaS, like getting rid of the web servers and going to Azure Websites.
Comments closedRegis Baccaro shows how to allow non-domain admins to configure Kerberos Constrained Delegation:
Now I need to add some special permissions to computer objects, so I click Add again. Once again, I’ll select the DBA group, then I need to switch to Descendant Computer objects. I click Write and then scroll down until I see Validated write to service principal name. I’ll click the box to enable it, and then OK, OK, and OK.
The end result looks like below :
2 permissions for DBA group,
All descendants objects : Write all properties
Descendant computer objects : Validate write to Service Principal Name
Regis has the whole process documented well, so check it out.
Comments closedBrent Ozar has a short test on effective RPO and RTO:
You can’t repair the corruption (it’s a clustered index, and there aren’t enough nonclustered indexes to cover all the columns), and the business needs that data back. You’re on: answer these questions:
Click through for the questions. They’re definitely worth thinking about.
Comments closedIt’s tempting to think, when p \ge \alphap≥α, that you’ve found the opposite thing from the p < \alphap<αcase: that you get to conclude that there is no statistically significant difference between the two averages. Don’t do that!
Simple statistical tests like the tt-test only tell you when averages are different; they can’t tell you when they’re the same. When they fail to find a difference, there are two possible explanations: either there is no difference or you haven’t collected enough data yet. So when a test fails, it could be your fault: if you had run a slightly larger experiment with a slightly larger NN, the test might have successfully found the difference. It’s always wrong to conclude that the difference does not exist.
It’s an interesting read. H/T Emmanuelle Rieuf.
Comments closedAsaf Yigal has a two-part series on comparing Apache Kafka and Redis for moving log events into Elasticsearch. Part 1 explains the technologies:
Redis is a bit different from Kafka in terms of its storage and various functionalities. At its core, Redis is an in-memory data store that can be used as a high-performance database, a cache, and a message broker. It is perfect for real-time data processing.
The various data structures supported by Redis are strings, hashes, lists, sets, and sorted sets. Redis also has various clients written in several languages which can be used to write custom programs for the insertion and retrieval of data. This is an advantage over Kafka since Kafka only has a Java client. The main similarity between the two is that they both provide a messaging service. But for the purpose of log aggregation, we can use Redis’ various data structures to do it more efficiently.
Part 2 compares the two technologies and explains which works better when:
Kafka heavily relies on the machine memory (RAM). As we see in the previous graph, utilizing the memory and storage is an optimal way to maintain a steady throughput. Its performance depends on the data consumption rate. In the case that consumers don’t consume data fast enough, Kafka will have to read from a disk and not from memory which will slow down its performance.
As you might expect, the answer for which technology to use is “it depends.”
Comments closedVarun Rao explains role-based access control using Apache Ranger on Amazon ElasticMapReduce:
Using the HUE SQL Editor, execute the following query.
These queries use external tables, and Hive leverages EMRFS to access the data stored in S3. Because HiveServer2 (where Hue is submitting these queries) is checking with Ranger to grant or deny before accessing any data in S3, you can create fine-grained SQL-based permissions for users even though there is a single EC2 role specified for the cluster (which is used by all requests the cluster makes to S3). For more information, see Additional Features of Hive on Amazon EMR.
If your job includes securing a Hadoop cluster, this is a nice read, even if you don’t use EMR.
Comments closedShane O’Neill learns more about SET STATISTICS IO’s scan count:
My previous intuitions about this guy was…
“Scan count is the number of hits the table/index received”
(THIS IS NOT CORRECT! …and why it isn’t down as a full quote).Let’s check out the definition again and see what it has to say for itself.
This was a fun read.
Comments closedPeter Schott extends a script to clean up the SSIS catalog database:
I really appreciate what MS has done w/ the SSIS Catalog. We have built-in logging at a level that wasn’t possible in prior releases, but that comes at a cost. The default retention is pretty high (365 days) and MS chose to handle cleanup using cascading deletes. This sort of coding makes life easier for the developers, but almost always performs poorly. That’s especially the case when you have 1 parent row with 100’s of thousands or child rows related to it. The problem is compounded because the first time people realize they need to lower the retention limit is about the same time that the database is starting to fill up or has filled up. At that point, it’s too late to lower the retention by a large number because the corresponding delete will still cause issues.
Click through for a script which helps extricate you from sticky situations. The ideal scenario here would be to set your retention period correctly and not have to delete rows directly, but sometimes you’re stuck in a less-than-ideal situation.
Comments closedTodd McDermid is excited about data flow groups in Integration Services:
Data Flow Groups
Data Flow Groups is what they’re calling it, and it’s deceptively simple to use. One of the reasons I’m sure I (and SSIS people I talk to who DID NOT LET ME KNOW IT WAS THERE) missed it is because I was expecting it to be a component in the toolbox. Not so.Code up your Data Flow as you normally would. Then go and select the components that you want to group together – via clicking and dragging a selection window, or click-selecting components. Any component combinations you want. Then right-click and select Group.
I admit that I didn’t know it existed either. This does seem rather useful.
Comments closed