Task list for cloud computing application management

Moving apps to cloud computing is not the last thing. Sometimes there are unexpected events that force you to redesign your application, and compliance requirements can create barriers to development, and if your cloud provider does not support low-level network services such as multicast, it can cause bandwidth problems.

After you evaluate which applications can run in a public cloud, there are other factors to consider, such as configuration, data migration, and monitoring. What are the most common configuration tasks you need to keep in mind when moving an application to cloud computing? The following list lists several key points:

• Service catalogs may include older versions of virtual machines (VMS) that are unable to meet your needs because they are not upgraded. Therefore, you need to use a virtual machine version that has the latest security and feature patches.

• View all non-interactive daemons and services running on a virtual machine and close all unwanted programs and services. This will reduce the attack services for all virtual machines.

• Configure the firewall between application tiers to minimize network traffic between tiers. For example, you should only open ports that you need to accept, respond to requests, and support management protocols.

• If you need powerful encryption features such as PCI DSS, make sure your TLS encryption function is powerful. Clients and servers in an SSL session will negotiate which cipher suite to use, and they may not use strong encryption if they do not have the appropriate configuration set up.

• When using a hybrid cloud computing model, you should configure a virtual private network to ensure communication between internal and external resources.

• Use SSL server certificates to authorize the server and to support encrypted communication. Depending on your architecture, you may be able to use domain level or multihomed name (SAN) certificates to help minimize the number of certificates you have to manage.

• Use configuration tools such as Chef or puppet to automate cloud computing management tasks.

Ways to achieve data migration to cloud computing

In addition to configuring cloud computing resources, you need a plan to move data from existing systems to cloud-based systems. There are several ways to implement data migration, and one of the best choices depends on your enterprise's requirements and system architecture.

One way to do this is to use the database replication feature to create a copy of your database in the cloud before migrating the system. If you use cloud computing technology for disaster recovery tasks (DR), then a copy of the database is already in the cloud. With this method, you can write the changes to the copy to keep your production system running properly. Also, the time required to migrate from current production systems to cloud computing systems has been minimized.

Backup and recovery is another method of data migration, which is applicable to systems based on non-transaction business, such as data warehousing and other business intelligence (BI) databases for data mining applications. However, this method delays the initial loading time based on cloud data and incremental updates.

In addition, not all data is stored in a relational database. For file-based data storage, you may want to use rsync, a synchronization tool that is commonly used for mirroring and backup service applications. Rsync is suitable for UNIX and Windows platforms.

Efficiently monitor cloud-based applications

The application's cloud migration will expand your infrastructure and create additional resources that you must monitor and manage. In order for all applications to function correctly, you should:

• Do a good job including logging of cloud computing services in the central Log Management library. You may want to maintain a separate log management library in the cloud to reduce network traffic and changes between cloud computing and your internal central log management library. But remember, this violates the principle of accessing log data from a single point, creating problems.

• Use event triggering mechanisms to generate alerts related to important events in cloud computing servers.

• Periodically use vulnerability scanning tools such as tenable company Nessus to check the security vulnerabilities of cloud-based virtual machines.

• If you are using a paid-for-use cloud application model, you should implement automatic tracking across servers and close some of these servers when usage drops to predefined thresholds. When the demand exceeds the predefined threshold of usage, the server can be added dynamically.

• Use spot pricing examples to achieve cost savings. On demand, Amazon EC2 provides available resources at variable prices. For example, you can bid at a specific price per virtual machine per hour, as long as your bid is still higher than the set price, you can continue to use the virtual machine. This pattern can still be applied if you do not want to complete the task at a preset date or time. If the bid is below the spot price and your virtual machine shuts down, then the spot pricing model also needs to restore your work without losing too much work.

Check your system architecture before moving applications to cloud computing. Pay special attention to the network services and traffic between the internal server and potential cloud-based servers. When it makes sense to move some applications to cloud computing, carefully consider system configuration, data migration processes, and ongoing monitoring and management procedures.

(Responsible editor: The good of the Legacy)