Name: Pradyumna Anil Kumar Kubear
Title of the paper: The Google File System
Objective:
This article discusses the Google File System. The Google File System, an expandable
distributed file system, was designed and implemented by the authors. which is being
developed to meet Google's rapidly expanding data processing needs.
Assumptions and Predictions (Hardware and Software):
The authors used hardware such as storage machines and client machines. Aside from these,
they have made use of low-cost commodity hardware.
The authors have mentioned software contributions such as file system interface extensions,
which are primarily designed to support distributed applications, error detection, fault
tolerance, and constant monitoring.
Major Contribution
The authors have contributed to the field of large-scale data processing by developing a new
file system that is intended to support large-scale data processing.
The authors conducted extensive research on traditional file systems and compared their
proposed model to traditional file system approaches.
Authors designed this approach after conducting all necessary research.
Pros and Cons
The authors mentioned benefits such as fault tolerance, high aggregate throughput, and
reduced master involvement.
A) Fault Tolerance
The authors used a checksum approach to detect corruption at the disk or IDE level.
B) High aggregate throughput
By separating file system control from data transfer, Google File System provides a high
aggregate throughput to many readers and writers.
C) Reduced master involvement
The master's involvement in common functioning is minimized, reducing the master to a
simple centralized element that does not become an issue.
Cons
A) Researchers believe that in the coming days they will update the networking stack to
overcome the current limitation on write throughput.
B) This application is restricted to the Google File System and may not be applicable to data
processing tasks of similar magnitude and consciousness.
