Cloud and Virtualization Based Log Management Service
With our computers becoming more capable and internet becoming more robust, virtualization- and cloud-based services have become prominent in their use. Virtualization technology has become a cost-effective and fast way to run simulations of models. Appli
- PDF / 159,920 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 96 Downloads / 212 Views
Abstract With our computers becoming more capable and internet becoming more robust, virtualization- and cloud-based services have become prominent in their use. Virtualization technology has become a cost-effective and fast way to run simulations of models. Application virtualization helps run programs in environments different from the underlying OS. The cloud, similarly, has become an indispensable platform to deliver a variety of services, such as storage, software, network, infrastructure, and even virtualization. The aim is to integrate virtualization using open virtualization archives (OVAs) or LXC Containers with OpenStack [1] to host a cloud, along with real-time data analytics implanted using ELK (Elasticsearch [2, 3], Logstash, Kibana) stack [4] to create a platform which provides an alternate approach to the present practices of log management in companies, as the present log management technologies lack in one or more of efficiency, security, isolation, and ease of use.
⋅
⋅
⋅
Keywords Log management Cloud computing Virtualization technology OpenStack ELK stack Elasticsearch Logstash Kibana Infrastructure as a service
⋅
⋅
⋅
⋅
⋅
1 Introduction Companies employ a large number of machines, such as servers, routers, and Linux machines. These produce large amounts of log messages each day, which are stored in dedicated log files, by default. These logs need to be analyzed and diagnosed to identify problems in the log and event data. For this purpose, the companies have S.R. Ghanta ⋅ A. Mukherjee (✉) Computer Science Department, Birla Institute of Technology, Mesra, Jaipur 27 Malviya Industrial Area, Jaipur 302017, Rajasthan, India e-mail: [email protected] S.R. Ghanta e-mail: [email protected] © Springer Science+Business Media Singapore 2017 S.K. Sahana and S.K. Saha (eds.), Advances in Computational Intelligence, Advances in Intelligent Systems and Computing 509, DOI 10.1007/978-981-10-2525-9_21
211
212
S.R. Ghanta and A. Mukherjee
system admins. These system admins use tools like Cat, Tail, and Grep, which are popular tools for log management. While this present system works, it lacks in a few areas, and does not offer much efficiency. A newer system is proposed herein, which intends to counter these issues.
1.1
Problems in the Present System
The sheer volume of logs generated by every machine, each day makes this a complex task. Upon identifying problems in logs, they are often compared with logs from previous days and records to identify a common cause to the problem, or to identify a pattern in the errors, so that further errors can be predicted and prevented. However, there is no log management service that makes this process intuitive. The present log management solutions do not offer any way to cross-reference the present log record with previously encountered log records to make helpful analysis. The system admin generally has to scroll through the multitude of log messages, and manually search for that instance, which displayed similar characteristics. This is a time-consu
Data Loading...
