Today, disasters come in many forms. Destruction, theft, loss, or natural disasters can crash an enterprise ’s application and destroy its data. In an ideal situation, the enterprise's data protection infrastructure can immediately recover all applications and data at the point of failure.
Enterprises can immediately switch faulty applications and continuously copy their data to achieve near-zero losses. But these operations consume resources and are expensive. In fact, IT departments need to set different recovery time and recovery point goals based on budget, resources, and application priorities.
People call these two goals Recovery Time Objective (RTO) and Recovery Point Objective (RPO). They are related, and both are necessary for application and data recovery. They are also metrics for different purposes.
The following discusses what they are, their similarities and differences, and why it is necessary to analyze the priority of applications to balance resources and application availability.
|| Define RTO and RPO
(1) RTO: recovery time objective
RTO refers to how long an application can be interrupted or closed without causing significant damage to the business. Some applications may be down for several days without serious consequences. And some high-priority applications can only be stopped for a few seconds, otherwise it will make it difficult for enterprises and customers to cope, and cause business loss.
RTO is not just the duration between business loss and recovery. This goal also includes the steps that IT departments must take to restore applications and their data. If IT has invested in failover services for high-priority applications, they can safely express RTO in seconds (IT departments must restore the local environment, but because applications are being processed in the cloud, the IT department may need some time).
The RTO task of an enterprise is to classify applications according to priority and potential business losses, and to match the resources of the enterprise accordingly. For example, a typical plan for near-zero RTO will require failover services. The 4-hour RTO allows local recovery from bare metal recovery and ends with full application and data availability. For RTOs longer than 8 hours, the IT team can sign maintenance contracts with local system integrators.
(2) RPO: recovery point objective
The recovery point objective refers to the loss tolerance of the enterprise: the amount of data that may be lost before causing significant damage to the business. This goal is expressed as a time measurement from the loss event to the most recent previous backup.
If all or most of the data is backed up in regularly scheduled 24-hour increments, then in the worst case, the enterprise will lose 24 hours of data. This is acceptable for some applications, but not for others.
For example, if the enterprise's application has a 4-hour RPO, the maximum interval between backup and data loss will be 4 hours. Having 4 hours of RPO does not necessarily mean that companies will lose 4 hours of data. For example, if a word processing application stops running at midnight and fails in the early morning, then there may not be too much (or any) data lost. But if a busy application closes at 10 am and does not resume until 2 pm, then the enterprise may lose 4 hours of high-value and possibly irreplaceable data. In this case, more frequent backups are needed to access the application-specific RPO.
This depends on the application priority. The range of a single RPO is usually 24 hours, 12 hours, 8 hours, and 4 hours. Measured in seconds to near zero. As long as the impact on the production system is minimal, RPOs over 8 hours can take advantage of existing backup solutions. The 4 hour RPO will require scheduled snapshot replication, while the near zero RPO will require continuous replication. When both RPO and RTO are close to zero, continuous replication is used in conjunction with failover services to achieve nearly 100% application and data availability.
|| How RTO and RPO are similar and different reasons
(1) Several features of RTO and RPO
* Recovery time and recovery point targets vary according to application and data priority. Even the largest and most powerful companies cannot provide near-zero RTO or RPO for all applications and should not do so.
* The only way to ensure 100% uptime (RTO) and no data loss (RPO) is to invest in a failover virtual environment with continuous data replication.
* IT prioritizes the processing of applications and data to match the realized RTO and RPO fees. Please note that priorities depend not only on income, but also on risk. Enterprises may not use applications frequently, but if their data is regulated, data loss may result in huge fines.
* RTO and RPO are measured in units of time. For RTO, the metric is the amount of time between application failure and full availability, including data recovery. RPO is also measured in units of time. The metric is the time interval between data loss and the previous backup. For RTO and RPO, the application / data priority can be directly converted into shorter time units.
(2) There is a huge difference between the goals of RTO and RPO
Despite their similarities, RPO and RTO serve different goals. RTO involves applications and systems, but mainly describes the limitations of application downtime.
RPO is mainly related to the amount of data lost after a failed event. However, customer transactions that cost hundreds of thousands of dollars will be catastrophic.
|| Examples of RTO and RPO in action
• Single file recovery: For example, a company employee accidentally deletes a time-sensitive email, and then empties the contents of the recycle bin and folder. Because Microsoft Exchange is the company's business-critical application, the IT department continues to support incremental changes in Exchange. And because their backup applications are capable of detailed backup and recovery, they can restore individual files within 5 minutes of RTO instead of restoring the entire virtual machine for individual files.
• E-commerce website: For example, a retail store ’s self-operated e-commerce website uses three different databases: a relational database that stores product catalogs, a document database that reports historical order data, and an API database connected to its payment processor gateway. The file database can reconstruct data from other databases, so its RTO and RPO are within 24 hours. The business only adds products to the relational database once a week, so RPO is not important. The RTO is that if the database is closed, the customer transaction stops.
To maintain high availability, the store uses failover services, so the database is immediately running on the virtual server. The company replicated the few changes it made within a week to its provider's disaster recovery platform. The API database contains ordering information and it takes a few seconds to complete the RPO and RTO. The IT department continuously copies data to the failover site. If the API database goes down, the site will immediately take over the processing.
Cost consideration
The survey shows that companies with an annual revenue of $ 100 million will lose about $ 275,000 during a 24-hour downtime. In the 4-hour snapshot replication plan, about $ 45,000 will be lost, and using near-zero continuous replication will cost about $ 7,600.
In fact, this number may be smaller or larger, depending on the time of day and application activity of the enterprise. Busy tasks or business-critical applications will lose more data and higher-priority data than less frequent applications.
Enterprises need to plan RPO and RTO accordingly and purchase the required resources before they are needed. Just like buying insurance, companies may never have to use them, but they may save their business.
Computer Cable include computer power cord, hdmi cable, hdmi to hdmi cable,vga cable, digital coaxial cable , usb printer cable, mini hdmi cable, local communication cable, LAN cable, long distance symmetric cable and submarine cable etc.
Yidashun produces hdmi cable, VGA cable and printer cable mainly, the length is 1.5m, 3m, 5m, 10m, 15m, 20m, 25m, 30m, 40m, 50m for choose. Yidashun has 1080P and 4K both.
Computer Cable,Computer Power Cord,Mini Computer Cable,USB Computer Cable
Shenzhen Yidashun Technology Co., Ltd. , https://www.ydsadapter.com