I read an article titled “Architecting Cloud Scale Identity Fabric” about concept of Identity Management of a Service as part of EEDC assignment. This article is written by Eric Olden, from Symplified. The main thing about this article is about the need of a service to manage user identity in the cloud. Well, I think this diagram from Symplified website is worth more than thousand words:
CAP theorem is widely used in Distributed Database System(DDBS) design. In a nutshell, it says that in designing modern DDBS, we only can choose two properties out of three properties that are crucial for DDBS. The aforementioned properties are Consistency (C), Availability (A) and Partition Tolerance (P). And this diagram below summarize the available combination of CAP:
Well, it has been more than a week since my last post. I was away on Andalucia trip, visiting Sevilla, Granada, Cordoba and Malaga. And here are some highlight of the trip:
Awesome Alhambra is Awesome!
Yes2, Alhambra in Granada is cool! Among the castles in Sevilla, Cordoba, and Malaga, Alhambra is simply the best! It has the most sophisticated architecture and Generalife garden is awesome as well. If you want to visit it, book your ticket early so you do not need to queue in the morning. You will be allocated morning time or afternoon time (each around 5 hours) and optimized your time to take cool picture or simply to enjoy the beauty of Alhambra and Generalife.
Well, I would like to write something about our Decentralized System (DS) project and what we are going to do in this project. Since it is the first post about the big picture of our DS project in this blog, I named it “Pilot” :p. Our group is G007, which consists of Julia, Diego, Enkhjin and myself.
When I was looking for example of P2P distributed storage system, I came across video from Google Tech Talk about Wuala. Wuala is an example of distributed peer-to-peer storage system. It used to be startup company, but LaCiE bought it in 2009. It allows you to store your data in the cloud, set up online back up and access your files, share your files with your friends easily, and access them from other computer.
This time, our group needed to prepare presentation about Apache Flume for EEDC homework. Flume is intended to solve challenges in safely transferring huge set of data from a node (example: log files in company web servers) to data store (example: HDFS, Hives, HBase, Cassandra etc etc).
Well, for a simple system with relatively small data set, we usually customize our own solution to do this job, such as to create some script to transfer the log to database. However, this kind of ad-hoc solution is difficult to make it scalable because usually it is created very tailored into our system. It sometimes suffers from problem in manageability, especially when the original programmer or engineer who created the system left the company. It is also often difficult to extend and, furthermore it may have problem in reliability due to some bugs during the implementation.
Well, it’s not really about relationship status :p.. since “they” are referring to distributed systems 😀 #geek
But it is about my latest review for a paper titled titled “Exploiting Availability Prediction in Distributed Systems“, by James W. Mickens and Brian D Noble. As we all know, availability is one of the important properties of distributed systems. Availability is concerned with the capability of a distributed system to serving its client properly although there are some component failures inside the system.They argued that availability modeling is crucial to (generally) make the system better, in term of resource efficiency and understanding system-wide phenomenon. Therefore, they propose a new way to predict availability, and they applied the predictor to three case studies. They found that their predictor works well under test data and they successfully shows that good predictor can improve the systems. Continue reading How available are they?
Quick summary about volunteer computing for school assignment..
Volunter computing is a type of distributed computing in which computer owners donate their computing resources (such as processing power and storage) to one or more “projects”. Continue reading Volunteer Computing
Last week, I have another assignment of Execution Environment of Distributed Computing course to create an elevator pitch about REST. More specifically, it is a short presentation (only 5 minutes) to convince your audience to use REST instead of SOAP.
I’m a Juventus fan! And I’m lucky enough to watch them live on December last year. And here is the story of my trip to watch Juventus!
This is the 1st part of 3-part story that I plan to write. The 1st part is about the preparation before the trip. The 2nd part is about the match that I attended, and the last part is about Juventus Stadium Tour. :). And without further ado, here they are.. the 1st part about the trip preparation! Continue reading Juventus Trip – Preparation