Apache Spark vs. Google BigQuery vs. Oracle Exadata

Well suited: To most of the local run of datasets and non-prod systems - scalability is not a problem at all. Including data from multiple types of data sources is an added advantage. MLlib is a decently nice built-in library that can be used for most of the ML tasks. Less appropriate: We had to work on a RecSys where the music dataset that we used was around 300+Gb in size. We faced memory-based issues. Few times we also got memory errors. Also the MLlib library does not have support for advanced analytics and deep-learning frameworks support. Understanding the internals of the working of Apache Spark for beginners is highly not possible.

Incentivized

Event-based data can be captured seamlessly from our data layers (and exported to Google BigQuery). When events like page-views, clicks, add-to-cart are tracked, Google BigQuery can help efficiently with running queries to observe patterns in user behaviour. That intermediate step of trying to "untangle" event data is resolved by Google BigQuery. A scenario where it could possibly be less appropriate is when analysing "granular" details (like small changes to a database happening very frequently).

Incentivized

• First, get the database on Oracle. If you are in an Oracle stack, it would be much better to use the Oracle products. If you are driving a Ferrari, you wouldn’t put a Mercedes engine in it. If you are writing a query, you cannot rely on other brands. Since I'm an architect, when I look for a product, I look for performance.
• The installation is easy because it comes out-of-the-box and you just start using it.
• Previous to Oracle Exadata, we were using a normal Oracle RAC service. We were just waiting for this product to come out.
• I'm currently writing a data warehouse on Exadata. Before this solution, we were aiming for this to be completed by 8 a.m., when our ETLs would finish. With the help of Exadata's special features, this was reduced to 3 a.m. This solution allows us to bring more data within the same time period. It provides us with more subject areas that provide more reports to our users. Our ETL times reduced to 65%, then to 50%.

Incentivized

• Rich APIs for data transformation making for very each to transform and prepare data in a distributed environment without worrying about memory issues
• Faster in execution times compare to Hadoop and PIG Latin
• Easy SQL interface to the same data set for people who are comfortable to explore data in a declarative manner
• Interoperability between SQL and Scala / Python style of munging data

Incentivized

• Realtime integration with Google Sheets.
• GSheet data can be linked to a BigQuery table and the data in that sheet is ingested in realtime into BigQuery. It's a live 'sync' which means it supports insertions, deletions, and alterations. The only limitation here is the schema'; this remains static once the table is created.
• Seamless integration with other GCP products.
• A simple pipeline might look like this:-
• GForms -> GSheets -> BigQuery -> Looker
• It all links up really well and with ease.
• One instance holds many projects.
• Separating data into datamarts or datameshes is really easy in BigQuery, since one BigQuery instance can hold multiple projects; which are isolated collections of datasets.

Incentivized

• High speed of SQL operations due to a unique design of Exadata with offloading of SQL processing to storage cells
• Built-in High Availability of a DB server due to it's base architecture of a multi-node Oracle Real Application Cluster
• High overall sever performance due to its use of a proprietary "Smart Cache" feature utilizing a high speed flash memory
• Excellent scalability of a DB server by adding cluster nodes as well as expanding it into a network of serially connected clusters

Incentivized

• Memory management. Very weak on that.
• PySpark not as robust as scala with spark.
• spark master HA is needed. Not as HA as it should be.
• Locality should not be a necessity, but does help improvement. But would prefer no locality

Incentivized

• Please expand the availability of documentation, tutorials, and community forums to provide developers with comprehensive support and guidance on using Google BigQuery effectively for their projects.
• If possible, simplify the pricing model and provide clearer cost breakdowns to help users understand and plan for expenses when using Google BigQuery. Also, some cost reduction is welcome.
• It still misses the process of importing data into Google BigQuery. Probably, by improving compatibility with different data formats and sources and reducing the complexity of data ingestion workflows, it can be made to work.

Incentivized

• Integrating with different types of databases can be a challenge
• Having a hardware platform optimized to run specific database software surely comes with a price
• Single vendor to support same hardware and software will make migrating to different hardware and/or software become a hassle

Incentivized

Capacity of computing data in cluster and fast speed.

We have to use this product as its a 3rd party supplier choice to utilise this product for their data side backend so will not be likely we will move away from this product in the future unless the 3rd party supplier decides to change data vendors.

Incentivized

If the team looking to use Apache Spark is not used to debug and tweak settings for jobs to ensure maximum optimizations, it can be frustrating. However, the documentation and the support of the community on the internet can help resolve most issues. Moreover, it is highly configurable and it integrates with different tools (eg: it can be used by dbt core), which increase the scenarios where it can be used

Incentivized

I think overall it is easy to use. I haven't done anything from the development side but an more of an end user of reporting tables built in Google BigQuery. I connect data visualization tools like Tableau or Power BI to the BigQuery reporting tables to analyze trends and create complex dashboards.

Incentivized

Excellent machine for your database needs . Don’t have to think twice if you have the budget to own it

Incentivized

I have never had any significant issues with Google Big Query. It always seems to be up and running properly when I need it. I cannot recall any times where I received any kind of application errors or unplanned outages. If there were any they were resolved quickly by my IT team so I didn't notice them.

Incentivized

I think Google Big Query's performance is in the acceptable range. Sometimes larger datasets are somewhat sluggish to load but for most of our applications it performs at a reasonable speed. We do have some reports that include a lot of complex calculations and others that run on granular store level data that so sometimes take a bit longer to load which can be frustrating.

Incentivized

1. It integrates very well with scala or python. 2. It's very easy to understand SQL interoperability. 3. Apache is way faster than the other competitive technologies. 4. The support from the Apache community is very huge for Spark. 5. Execution times are faster as compared to others. 6. There are a large number of forums available for Apache Spark. 7. The code availability for Apache Spark is simpler and easy to gain access to. 8. Many organizations use Apache Spark, so many solutions are available for existing applications.

BigQuery can be difficult to support because it is so solid as a product. Many of the issues you will see are related to your own data sets, however you may see issues importing data and managing jobs. If this occurs, it can be a challenge to get to speak to the correct person who can help you.

Incentivized

Spark in comparison to similar technologies ends up being a one stop shop. You can achieve so much with this one framework instead of having to stitch and weave multiple technologies from the Hadoop stack, all while getting incredibility performance, minimal boilerplate, and getting the ability to write your application in the language of your choosing.

Incentivized

PowerBI can connect to GA4 for example but the data processing is more complicated and it takes longer to create dashboards. Azure is great once the data import has been configured but it's not an easy task for small businesses as it is with BigQuery.

Incentivized

Oracle Exadata Database Machine had the best performance overall hands down. It clearly beat the competition and we were seeing 1000X improvement on SAP HANA. Oracle Exadata Database Machine beat that without us refactoring our code. To achieve that in HANA, we had to refactor the code somewhat. Now this was for our limited POC of 5 use cases. Given the large number of stored procedures we had in Sybase, we need to capture more production metrics but we are seeing incredible performance.

Incentivized

None so far. Very satisfied with the transparency on contract terms and pricing model.

We have continued to expand out use of Google Big Query over the years. I'd say its flexibility and scalability is actually quite good. It also integrates well with other tools like Tableau and Power BI. It has served the needs of multiple data sources across multiple departments within my company.

Incentivized

Google Support has kindly provide individual support and consultants to assist with the integration work. In the circumstance where the consultants are not present to support with the work, Google Support Helpline will always be available to answer to the queries without having to wait for more than 3 days.

• Business leaders are able to take data driven decisions
• Business users are able access to data in near real time now . Before using spark, they had to wait for at least 24 hours for data to be available
• Business is able come up with new product ideas

Incentivized

• Previously, running complex queries on our on-premise data warehouse could take hours. Google BigQuery processes the same queries in minutes. We estimate it saves our team at least 25% of their time.
• We can target our marketing campaigns very easily and understand our customer behaviour. It lets us personalize marketing campaigns and product recommendations and experience at least a 20% improvement in overall campaign performance.
• Now, we only pay for the resources we use. Saved $1 million annually on data infrastructure and data storage costs compared to our previous solution.

Incentivized

• Single support from a single vendor with both machine and database from Oracle, which is costing us less.
• With Exadata, we need less technical manpower and less technical support. A business transaction with the integrated and centralized database helps us focus on other business needs.
• We don't need to buy additional licenses and Hardware for the next 3 to 5 years.

Incentivized