Why Mission Critical Cloud App Breakdowns Require Lengthy Fixes

With most enterprises migrating at least one-quarter of their highly available (HA) and mission critical apps to the cloud, organizations are finding that app performance issues have emerged as regular occurrences, according to a recent survey from SIOS Technology. The resulting report, titled “The State of Application High Availability,” reveals that repairs for cloud-based HA apps—such as order-taking systems and customer relationship management (CRM) tools that keep business going—often involve multiple performance monitoring tools and report. They also require several hours (or longer) to resolve. A total of 390 IT professionals and decision-makers took part in the research, which was conducted by ActualTech Media. The following slideshow presents survey highlights, with charts provided courtesy of SIOS Technology, which makes software for optimizing and maintaining HA applications.

SIOS reports that 56 percent of survey respondents said at least one-quarter of their highly available (HA) apps are currently in the cloud. Nearly one-quarter said more than one-half of their HA apps are in the cloud.

Among those whose organizations that haven’t fully moved their HA apps to the cloud, 60 percent said it’s because they prefer to keep HA apps on-premise to maintain more control over the infrastructure. And 22 percent said their organization lacks the internal knowledge/expertise to make this shift.

More than seven of ten respondents said they run into an app performance issue that impacts end users at least monthly. More than one-third said this happens no less than two to three times a week, if not daily.

When there is a performance issue involving an app in the cloud, more than four of five respondents said they need to consult with at least two to four tools (such as monitoring apps, reports and analytics tools) to identify the cause. More than one-in-ten said they consult with at least five such tools.

The report indicates that 77 percent of respondents said it takes at least one to three hours to resolve an app performance issue. More than one-in-five said it takes no less than three hours – if not five or more.

App or database issues cause most performance issues for HA apps in the cloud, as cited by 64 percent of respondents. The cloud platform provider in question ranks second (as cited by 17 percent), followed by infrastructure issues (15 percent).

To ensure the high availability of mission-critical apps in the cloud, 71 percent of respondents said they resort to redundancy and/or the elimination of single points of failure (a.k.a. clustering). About one-quarter overprovision and/or outsource HA app oversight to a managed service provider.

The majority of respondents – 56 percent – rely upon Oracle for their apps/database environment. SQL Server ranked second, as cited by 49 percent of respondents.

In selecting SQL Server editions, 58 percent of respondents said their organization is running Enterprise. About one-quarter of companies are running Standard and Business Intelligence.

The findings reveal that 86 percent of respondents said their organization runs at least one database in a high availability/clustered environment. This figure remains pretty consistent among respondents regardless of their employer’s size, as 88 percent of respondents from small companies reported this, along with 82 percent of those at medium-sized and 86 percent of those at large-sized organizations.