Pentaho is a suite of open source business intelligence and analytics products, now offered and supported by Hitachi Data Systems since the June 2015 acquisition.
Apache Pig and Apache Hive provide most of the things spark provide but apache spark has more features like actions and transformations which are easy to code. Spark uses optimization technique as we can select driver program and manipulate DAG (Directed Acyclic Graph) Python …
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.
Pentaho is very well suited to perform data extraction & data mining from various cloud storage & transform that data using various available data models. However, the software struggles when it comes to visualizing the extracted data in an appealing manner & can be difficult for end-users to get an understanding of data tables created using those models.
I think the relative obscurity of the tool is a downside, not as many developers, consultants or peers you can tap into.
Lack of a solid user community held us back, looking at Power BI and Qlik, they have huge user communities that help each other out. Would have liked that here.
Smaller company means smaller sales force, and the lack of a local presence made it hard to only interact online with the account rep. Other companies have someone local who often stops by with pre-sales developers to just pitch in free of charge when they have time.
I will use Pentaho until I find a better tool with a better, easier to use report designer client. For now, Pentaho has been the most powerful reporting tool for our clients because of its ability to connect to Odoo, integrate in Odoo (reports are accessible in Odoo) and the flexibility in report design and parameter integration
The only thing I dislike about spark's usability is the learning curve, there are many actions and transformations, however, its wide-range of uses for ETL processing, facility to integrate and it's multi-language support make this library a powerhouse for your data science solutions. It has especially aided us with its lightning-fast processing times.
The Pentaho tools are designed so you can start playing around on your own. Of course, you will need guidance at some point, but the training teams are good at guiding new users, and the online documentation is usually pretty up-to-date.
Some of the tools, such as the Pentaho Data Integration tool and the Pentaho Server, are pretty self-explanatory. The other tools maybe are not so quickly and obvious to use, but again, with some documentation and some customer support, you can find your way around them.
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.
They were responsive to our questions when we raised issues. They gave us workarounds when required. They were quite knowledgeable when it came to issue analysis and providing fixes. They were forthright in informing us if a bug was not due for release soon.
Course Taken: DI1000 Pentaho Data Integration Fundamentals Setup A week before your class started, the instructor will start sending out class material and lab setup instructions. This is helpful so that you understand how the environment is laid out and can start reviewing the content. Ultimately it saved about a 1/2 day trying to setup with 10 other people online which was great! The Course The 3-day course was laid out like many other technical classes with 15-30 minutes instruction and 15-60 minutes of lab exercises. The instructor was very knowledgeable with the functionality from version to version and answered questions as we went along. I was amazed at some of the functionality that was available that I was not using at the time and quickly implemented changes to many existing transformations and jobs. The novice users seemed to catch on quickly and more experienced users explained how some of the functionality was used in their home environments. Towards the end there was enough time so that we were able to ask very directed questions about our own environments. Overall, I really found the class to be informative and deliver enough information to be dangerous. My skills improved and I was able to design better and efficient transformations for the HIE. Course Description: https://training.pentaho.com/instructor-led-training/pentaho-data-integration-fundamentals-di1000
Get the right people in before starting implementation. Start small and build as you go approach is time consuming and involves lot of rework. Evangalize within the organization the capabilities and limitations equally so that correct delivery expectations are set. Set expectations with the Customer that the tool cannot replace proprietary software in terms of stability/usability and that timelines could change given the new ness of the product.
All the above systems work quite well on big data transformations whereas Spark really shines with its bigger API support and its ability to read from and write to multiple data sources. Using Spark one can easily switch between declarative versus imperative versus functional type programming easily based on the situation. Also it doesn't need special data ingestion or indexing pre-processing like Presto. Combining it with Jupyter Notebooks (https://github.com/jupyter-incubator/sparkmagic), one can develop the Spark code in an interactive manner in Scala or Python
Since the Pentaho platform offers a range of broad functionality across data preparation and advanced analytics, it also can be easily integrated to support many data sources and machine-learning frameworks. Based on that fact, we selected Pentaho to be used in our internal department. It also supports many of our BI use cases as required by company management or the business user. Last but not least, the Pentaho license is cheaper than their competitor.
Faster turn around on feature development, we have seen a noticeable improvement in our agile development since using Spark.
Easy adoption, having multiple departments use the same underlying technology even if the use cases are very different allows for more commonality amongst applications which definitely makes the operations team happy.
Performance, we have been able to make some applications run over 20x faster since switching to Spark. This has saved us time, headaches, and operating costs.
