Apache CouchDB vs. Qubole

Great for REST API development, if you want a small, fast server that will send and receive JSON structures, CouchDB is hard to beat. Not great for enterprise-level relational database querying (no kidding). While by definition, document-oriented databases are not relational, porting or migrating from relational, and using CouchDB as a backend is probably not a wise move as it's reliable, but It may not always be highly available.

Incentivized

I find Qubole is well suited for getting started analyzing data in the cloud without being locked in to a specific cloud vendor's tooling other than the underlying filesystem. Since the data itself is not isolated to any Qubole cluster, it can be easily be collected back into a cloud-vendor's specific tools for further analysis, therefore I find it complementary to any offerings such as Amazon EMR or Google DataProc.

Incentivized

• It can replicate and sync with web browsers via PouchDB. This lets you keep a synced copy of your database on the client-side, which offers much faster data access than continuous HTTP requests would allow, and enables offline usage.
• Simple Map/Reduce support. The M/R system lets you process terabytes of documents in parallel, save the results, and only need to reprocess documents that have changed on subsequent updates. While not as powerful as Hadoop, it is an easy to use query system that's hard to screw up.
• Sharding and Clustering support. As of CouchDB 2.0, it supports clustering and sharding of documents between instances without needing a load balancer to determine where requests should go.
• Master to Master replication lets you clone, continuously backup, and listen for changes through the replication protocol, even over unreliable WAN links.

Incentivized

• From a UI perspective, I find Qubole's closest comparison to Cloudera's HUE; it provides a one-stop shop for all data browsing and querying needs.
• Auto scaling groups and auto-terminating clusters provides cost savings for idle resources.
• Qubole fits itself well into the open-source data science market by providing a choice of tools that aren't tied to a specific cloud vendor.

Incentivized

• NoSQL DB can become a challenge for seasoned RDBMS users.
• The map-reduce paradigm can be very demanding for first-time users.
• JSON format documents with Key-Value pairs are somewhat verbose and consume more storage.

Incentivized

• Providing an open selection of all cloud provider instance types with no explanation as to their ideal use cases causes too much confusion for new users setting up a new cluster. For example, not everyone knows that Amazon's R or X-series models are memory optimized, while the C and M-series are for general computation.
• I would like to see more ETL tools provided other than DistCP that allow one to move data between Hadoop Filesystems.
• From the cluster administration side, onboarding of new users for large companies seems troublesome, especially when trying to create individual cluster per team within the company. Having the ability to debug and share code/queries between users of other teams / clusters should also be possible.

Incentivized

Because our current solution S3 is working great and CouchDB was a nightmare. The worst is that at first, it seemed fine until we filled it with tons of data and then started to create views and actually delete.

Incentivized

Personally, I have no issues using Amazon EMR with Hue and Zeppelin, for example, for data science and exploratory analysis. The benefits to using Qubole are that it offers additional tooling that may not be available in other cloud providers without manual installation and also offers auto-terminating instances and scaling groups.

Incentivized

Couchdb is very simple to use and the features are also reduced but well implemented. In order to use it the way its designed, the ui is adequate and easy. Of course, there are some other task that can't be performed through the admin ui but the minimalistic design allows you to use external libraries to develop custom scripts

Incentivized

it support is minimal also hw requirements. Also for development, we can have databases replicated everywhere and the replication is automagical. once you set up the security and the rules for replication, you are ready to go. The absence of a model let you build your app the way you want it

Incentivized

It has been 5+ years since we chose CouchDB. We looked an MongoDB, Cassandra, and probably some others. At the end of the day, the performance, power potential, and simplicity of CouchDB made it a simple choice for our needs. No one should use just because we did. As I said early, make sure you understand your problems, and find the right solution. Some random reading that might be useful: http://www.julianbrowne.com/article/viewer/brewers-cap-theorem https://www.couchbase.com/nosql-resources/why-nosql\ https://www.infoq.com/articles/cap-twelve-years-later-how-the-rules-have-changed

Incentivized

Qubole was decided on by upper management rather than these competitive offerings. I find that Databricks has a better Spark offering compared to Qubole's Zeppelin notebooks.

Incentivized

• It has saved us hours and hours of coding.
• It is has taught us a new way to look at things.
• It has taught us patience as the first few weeks with CouchDB were not pleasant. It was not easy to pick up like MongoDB.

Incentivized

• We like to say that Qubole has allowed for "data democratization", meaning that each team is responsible for their own set of tooling and use cases rather than being limited by versions established by products such as Hortonworks HDP or Cloudera CDH
• One negative impact is that users have over-provisioned clusters without realizing it, and end up paying for it. When setting up a new cluster, there are too many choices to pick from, and data scientists may not understand the instance types or hardware specs for the datasets they need to operate on.

Incentivized