We need a reasonable insert and query speed over huge tables so I considered using some noSQL adapter with Django. Unfortunately:
Django does not provide official support for noSQL databases.
In our original schema some Big Data are relational to other Big Data making the data duplication unacceptable.
Project deadlines are enemies of hot stuff like this.
So, as far I can see, PostgreSQL should be the way to go for this scenario, right?!
Please let me know any other detail that may be relevant to this question!
Bonus to anyone that can point out some useful database techniques like database sharding...
Well, there is a fork of django project that uses MongoDb as the backend.You can read about it here . The Code on GitHub is here.You give some heads up, MongoDB is a NOSQL db that does support sharding and replication. So i think this might something that you are looking for.
Related
Can we directly use Elasticsearch as a primary database in Django? I have tried finding the solution or a way out but could not find any relevant information. Everywhere it is said that we can use Elasticsearch as a search engine over any other primary database. But as per my understanding Elasticsearch is a NoSQL Database, so there should be a way to use it as a Primary Database in a Django Project.
Please help, if someone has any idea about it.
The short answer is no.
SO already has an answer here and this is still valid: Using ElasticSearch as Database/Storage with Django
ES is not a ACID compliant
Indexing is not immediate so any kind of load would be an issue
It's very weakly consistent
Use it together with a proper database and it will help with real time searches, analytics, expensive queries etc. but treat it as derived data.
I know some ways like scrapy-djangoitem but as it has mentioned:
DjangoItem is a rather convenient way to integrate Scrapy projects with Django models, but bear in mind that Django ORM may not scale well if you scrape a lot of items (ie. millions) with Scrapy. This is because a relational backend is often not a good choice for a write intensive applications (such as a web crawler), specially if the database is highly normalized and with many indices.
So what is the best way to use scraped items in db and django models?
It is not about Django ORM but rather about the database you choose as backend. What it says is that if you are expecting to write millions of items to your tables, relational database systems might not be your best choice here (MySQL, Postgres ...) and it can be even worse in terms of performance if you add many indicies since your application is write-heavy (Database must update B-Trees or other structures for keeping index on every write).
I would suggest sticking with Postgres or MySQL for now and look for another solution if you start to have performance issues on database level.
Amazon said
NoSQL design requires a different mindset than RDBMS design. For an RDBMS, you can create a normalized data model without thinking about access patterns. You can then extend it later when new questions and query requirements arise. For DynamoDB, by contrast, you shouldn't start designing your schema until you know the questions it needs to answer. Understanding the business problems and the application use cases up front is absolutely essential.
It seems that I should design the tables after designing the product for efficient query cost.
But a product can be pivoted or be appended new features. In early stage, nobody knows where the product goes.
Is dynamodb suitable for growing or pivotable product?
In my opinion, the main benefit of Dynamo DB over other NoSQL solutions is that it is a managed database service. You pay for reads and writes and you never worry about scaling to handle larger data, more users. If you are doing a prototype or don't have technical know-how to setup a database server and host in the cloud it could be useful and cost effective. It has its limitations however so if you do have technical resources consider another open source NoSQL option.
I think that statement by Amazon is confusing and is probably more marketing than anything else. Use NoSQL in cases where your data is only accessed in distinct elements that do no have to be combined in a complex manner. It's also helpful if you don't have an exact schema defined because NoSQL doesn't require a hard set schema you can store any fields in a table and you can always add new fields. This is helpful when things are changing rapidly and you don't want to migrate everything as strictly as an RDBMS would require. If however you're going to have to run complex logic or calculations combining data from across tables you should use an RDBMS. You could use NoSQL for some data and and RDBMS for other data in a hybrid fashion but in that case you probably wouldn't want to use Dynamo DB because you'd want full ownership to set it up properly. Hope this helps I'm sure others have more to say and I welcome comments to help me refine my answer.
This is more of an architectural question than a technological one per se.
I am currently building a business website/social network that needs to store large volumes of data and use that data to draw analytics (consumer behavior).
I am using Django and a PostgreSQL database.
Now my question is: I want to expand this architecture to include a data warehouse. The ideal would be: the operational DB would be the current Django PostgreSQL database, and the data warehouse would be something additional, preferably in a multidimensional model.
We are still in a very early phase, we are going to test with 50 users, so something primitive such as a one-column table for starters would be enough.
I would like to know if somebody has experience in this situation, and that could recommend me a framework to create a data warehouse, all while mantaining the operational DB with the Django models for ease of use (if possible).
Thank you in advance!
Here are some cool Open Source tools I used recently:
Kettle - great ETL tool, you can use this to extract the data from your operational database into your warehouse. Supports any database with a JDBC driver and makes it very easy to build e.g. a star schema.
Saiku - nice Web 2.0 frontend built on Pentaho Mondrian (MDX implementation). This allows your users to easily build complex aggregation queries (think Pivot table in Excel), and the Mondrian layer provides caching etc. to make things go fast. Try the demo here.
My answer does not necessarily apply to data warehousing. In your case I see the possibility to implement a NoSQL database solution alongside an OLTP relational storage, which in this case is PostgreSQL.
Why consider NoSQL? In addition to the obvious scalability benefits, NoSQL offer a number of advantages that probably will apply to your scenario. For instance, the flexibility of having records with different sets of fields, and key-based access.
Since you're still in "trial" stage you might find it easier to decide for a NoSQL database solution depending on your hosting provider. For instance AWS have SimpleDB, Google App Engine provide their own DataStore, etc. However there are plenty of other NoSQL solutions you can go for that have nice Python bindings.
i'm new to the nosql world, and from forums and articles that i've read: most of users try to "mix" nosql tools, for example, they use Cassandra and MongoDB together to make a "powerful system", because am beginning with MongoDB, i've downloaded the DjanMon project (am a django fan ^_^ ), of course i've downloaded the special version of django that accepts the NoSql use: Django NonRel, and i've noticed that the Setting file dont "oblige" you to use one specific NoSql solution like in Django with RDBMS where you must specify MySql or PostegreSql or other solution, so, is it possible to mix lot of (or two of course) NoSql solution using Django (for example MongoDB+Cassandra)?
There's nothing to stop you using multiple storage solutions, whether SQL or NoSQL - but the NoSQL solutions all have different architectures, data models and APIs (For example, MongoDB is a document-oriented database, whereas Cassandra is Column-oriented), so you can't usually swap one for another without some effort.
Can you clarify what you are actually trying to achieve? I.e. why are you interested in mixing these two specific solutions?