Regaining cloud
control with intelligent,
AI-powered observability

Com­put­ing envi­ron­ments are becom­ing increas­ing­ly labyrinthine, as busi­ness­es adopt hun­dreds or even thou­sands of cloud ser­vices, often select­ed by cen­tral IT or siloed depart­ments. The prob­lem is wors­ened by inte­gra­tion efforts that lag far behind rapid, ongo­ing inno­va­tion, includ­ing the reg­u­lar addi­tion of cus­tomer-fac­ing apps and func­tion­al­i­ty.

Six in 10 chief infor­ma­tion offi­cers (CIOs) expect that this dig­i­tal trans­for­ma­tion will con­tin­ue to accel­er­ate. Most IT envi­ron­ments are now mul­ti-cloud and change by the minute, with code qual­i­ty also slip­ping giv­en the con­stant pres­sure for inno­va­tion. Alarm­ing­ly, some 63% of CIOs now say their hybrid and mul­ti-cloud set­up is so com­plex that no human team could man­age it, accord­ing to the research by the soft­ware intel­li­gence com­pa­ny Dyna­trace.

“The need to con­stant­ly inno­vate, and to trans­form employ­ee and cus­tomer expe­ri­ences, means sys­tems are get­ting more deeply com­plex, and in essence, impos­si­ble for many com­pa­nies to man­age,” says Alois Reit­bauer, chief tech­nol­o­gy strate­gist at Dyna­trace. “Busi­ness­es are risk­ing a real loss of tech con­trol as they drown in the shift­ing sea of cor­po­rate tech­nol­o­gy.”

Prob­lems emerg­ing for busi­ness­es include unex­plained process fail­ure, new deploy­ments caus­ing laten­cies and out­ages, and ser­vice users not being able to com­plete trans­ac­tions. That is aside from dam­ag­ing secu­ri­ty prob­lems, with 71% of secu­ri­ty heads recent­ly warn­ing they are not ful­ly con­fi­dent their codes are free from vul­ner­a­bil­i­ties before going live. All these prob­lems can quick­ly erode busi­ness rev­enue and rep­u­ta­tion.

Com­pa­nies’ typ­i­cal response to these chal­lenges include intro­duc­ing appli­ca­tion per­for­mance mon­i­tor­ing (APM), but this often results in far more warn­ings being gen­er­at­ed than can be addressed. The aver­age num­ber of cor­po­rate secu­ri­ty alerts pre­sent­ed by an APM-based approach is 2,168 per month, and sev­en in 10 CIOs say their teams are left sub­merged in man­u­al tasks as a result.

“Busi­ness­es end up with a del­uge of infor­ma­tion that they don’t know what to do with,” Reit­bauer explains. “Their mon­i­tor­ing might high­light tens or even hun­dreds of sys­tems con­nect­ed to any one prob­lem, mean­ing a huge amount of work then has to be done to find where the root cause lies and how to fix it, a sit­u­a­tion wors­ened when there are mul­ti­ple oth­er prob­lems tak­ing place simul­ta­ne­ous­ly.” Many busi­ness­es per­sist with lega­cy mon­i­tor­ing tech­nol­o­gy, hav­ing stitched togeth­er up to 10 mon­i­tor­ing sys­tems, on aver­age deliv­er­ing observ­abil­i­ty into only 11% of their IT infra­struc­ture.

In addi­tion, the infor­ma­tion most busi­ness­es find on their sys­tems lacks con­text. “It’s a bit like hav­ing a clin­i­cian check someone’s tem­per­a­ture, find­ing it’s high, but not being giv­en any indi­ca­tions as to events that might reveal the under­ly­ing cause,” Reit­bauer notes. “They’d have to ask all sorts of ques­tions to get the broad­er pic­ture and find out what the prob­lem is.” Such ques­tions often then fail, because bias­es in human think­ing typ­i­cal­ly mean a focus on instinct or recent expe­ri­ences, which can lead even the sharpest IT pro­fes­sion­als to over­look the real caus­es.

As a result, many busi­ness­es are now turn­ing to Dynatrace’s auto­mat­ic and intel­li­gent observ­abil­i­ty, which has arti­fi­cial intel­li­gence for IT oper­a­tions (AIOps) at the core. Dynatrace’s uni­fied plat­form is easy to use and rapid­ly assess­es a range of pos­si­ble ques­tions, analy­ses what is hap­pen­ing across com­plex cloud envi­ron­ments in full con­text and assess­es user expe­ri­ence. It sifts through the infor­ma­tion to derive clear answers and pri­ori­tise urgent reme­di­al action. In the case of secu­ri­ty risks, sus­pi­cious actions are imme­di­ate­ly blocked. “The tech­nol­o­gy is designed for the veloc­i­ty, vol­ume and com­plex­i­ty of mod­ern cloud envi­ron­ments, and it is trans­par­ent and unbi­ased. It pro­vides clear, fac­tu­al expla­na­tions and action­able pri­or­i­ties based on busi­ness impact,” Reit­bauer explains.

A retail busi­ness expe­ri­enc­ing slow tech­nol­o­gy per­for­mance since a new app deploy­ment, for exam­ple, can use Dyna­trace to find exact­ly where and how prob­lems are grow­ing, and deter­mine how teams can opti­mise the user expe­ri­ence. Mean­while, a finan­cial firm can detect emerg­ing flaws in user expe­ri­ence and iden­ti­fy reme­dies, help­ing teams tran­si­tion from reac­tive to proac­tive. And a life sci­ences organ­i­sa­tion with IT secu­ri­ty con­cerns can visu­alise poten­tial impact and enable teams to col­lab­o­rate in response.

As IT infra­struc­tures become ever more sprawl­ing, it is time for busi­ness­es to imple­ment intel­li­gent observ­abil­i­ty and regain con­trol of com­plex cloud envi­ron­ments. They can do so by mov­ing from a del­uge of alerts to a uni­fied plat­form-based approach. This enables teams to auto­mat­i­cal­ly iden­ti­fy root caus­es, resolve issues quick­ly, and reduce time spent on man­u­al tasks so they can pri­ori­tise inno­va­tion.

To learn more about how Dyna­trace can help your busi­ness, vis­it dynatrace.com/trial and fol­low us on Twit­ter @dynatrace

Q&A: With companies drowning in the complexity of their systems, Alois Reitbauer, chief technology strategist at Dynatrace, explores how businesses can regain control
today

What are the dai­ly frus­tra­tions that you see IT, CloudOps and DevOps teams expe­ri­enc­ing with­out reli­able observ­abil­i­ty?

Often peo­ple don’t realise the prob­lems they are hav­ing in their apps. They may not even know that users can’t log in or that an app isn’t work­ing prop­er­ly. When they do iden­ti­fy a prob­lem, they may not under­stand it among the thou­sands of data points in front of them. This effec­tive­ly means they are los­ing con­trol of the tech­nol­o­gy they use.

Do com­pa­nies face a chal­lenge choos­ing the right observ­abil­i­ty sys­tem?

Yes. Typ­i­cal­ly, com­pa­nies have acquired sev­er­al appli­ca­tion per­for­mance mon­i­tor­ing sys­tems that don’t inter­act or pro­vide rapid­ly action­able data in con­text. Instead, it’s bet­ter to start with desired out­comes and ask: “How should prob­lems be iden­ti­fied and resolved?” Usu­al­ly, this will lead to intel­li­gent root cause and impact analy­sis, with respons­es auto­mat­i­cal­ly pri­ori­tised. This is where smart, AI-pow­ered observ­abil­i­ty comes in.

What are the ben­e­fits of intel­li­gent observ­abil­i­ty?

Auto­mat­ed prob­lem iden­ti­fi­ca­tion and res­o­lu­tion mas­sive­ly improve deci­sion mak­ing while sav­ing a great deal of time and mon­ey. AI and automa­tion should be high on every company’s pri­or­i­ty list. As more busi­ness­es get observ­abil­i­ty right, it will grad­u­al­ly become the norm for devel­op­ers to be able to sim­ply cre­ate and run tech­nol­o­gy that works pow­er­ful­ly and reli­ably, and to quick­ly reme­di­ate prob­lems when they arise.

Pro­mot­ed by Dyna­trace