As a child or adult, one question that have definitely aroused in everyone's mind is "Is this universe infinite or it has some boundaries?" as a science or astronomy student this might be the first question to arouse in students mind. Coming back to the answer of this beautiful question- The observable universe is still huge but it has limits. This is because we know that the universe isn't infinitely old - we know that the big bang occured some 13.8 billion years ago. So from this we can conclude that the light has had only 13.8 billion years to travel and hence defines the limitations of universe. All we know or scientists believe is that universe is still far above from our observable distance but at some point it has limits for sure.
As a beginners in field of machine learning and data science must be unaware of KIRA, so first question in your mind would be "what is KIRA?". so here is the answer to your question.
What is KIRA?
What is KIRA?
KIRA is a powerful machine learning software that identifies, abstracts and analyse text in your contracts and other documents.
How KIRA helps?
with over thousands of in built model, KIRA provides insights into various projects including:
- Due deligence
- least abstraction
- compliance
- finance
- deal points
How KIRA actually works?
KIRA extracts and automatically highlights the provisions that are important to you and helps in organizing your data for analysis.
Now after studying a lot about KIRA we are back to our topic i.e using KIRA as a tool for big data technology.
KIRA as a tool for big data technology
Scientific analysis most commonly compose multiple single process programs. Also we should
have knowledge thatan end to end dataflow of single process programs is called many task application. generally, HPC tools are used to parallelize these analysis. In this work, we use an alternate method that uses ApacheSpark to parallelize these analysis.
Here we implement KIRA, a flexible and distributed astronomy image processing toolkit, and its source extractor (KIRA-SE) application. Using KIRA SE , we examine various properties like flexibility, scheduling capacity, dataflow richness and performance of Apache Spark runing on the Amazon EC2 cloud. By exploiting data locality, KIRA SE achieves a 4.1 speedup over a equivalent c- program when analyzing a 1TB dataset using 512 cores on the Amazon EC2 cloud.
Further, it achieves a 1.8 speedup over the c program on NERSC Edison supercomputer.
So our final conclusion after our experience with KIRA demonstrates that data intensive computing platforms like Apache Spark are a good and performant alternative for many-task scientific applications.
Comments
Post a Comment