[PDF] [PDF] Army Tactical Mission Application Performance - Cisco AppDynamics

View - Download [PDF] Army Tactical Mission Application Performance - Cisco AppDynamics

Previous PDF

Next PDF

2 2 2

AppDynamics Application Intelligence

Transform the way modern applications are built,

maintained and operated Optimize user experience with continuous, mission-focused application eydCyalmieAaCmP|CyPmrA Approved for release within the Department of Army. Distribution is u nlimited.

Rick Noonan

Cisco

AppDynamics Specialist

US Army

Rick.Noonan@AppDynamics.com

(703)484-5671 AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

Abstract

|2Pag2eApA2 Dyn2am|2mimcPgi2Ps2mC2oCfDgm|oCdtn2iodoPmt2mdgl2Pag2oC.g |PygCP2oC2mCi2DgtomCfg2sC2 nl2r|gD|2DgtnoCd2sC2Pag|g2 o|agi2yo||osC2gLLgfPo.gCg||A2 sDdmCo6mPosC|2fDoPofo6g2gmfa2sPagDl2sLPgC2mCgfisPmttn2mCi24oPasrP2ygPDof|A22vaDsrdasrP2Pag2 Dynl2 s2Pgmy|2Dg|csC|ohtg2

LsD2Pao|2mDgm2mCi2

sLPgC2

4oPasrP2fsC|oigDmPosC2LsD2mtt2

Pag2fsyctg©oPn2PamP27v2hDoCd|A222

2 vag2ep2 Dyn8|2919 05 Ai rborne

Division

has been experiencing many of these daily, weekly, and |osC2mcctofmPosC|A22vag2 919
05 Airborne invited the Cisco AppDynamics team to participate in a recent e xercise in January I1I12 Ps2 g.mtrmPg2 2 cgDLsDymCfg

A22MrD2LoCioCd|2

mDg2hm|gi2sC24amP24g2sh|gD.gi2 r|oCd2 ymfaoC oCd 2 tgmDC oCd2oCagDgCP2oC2srD2|strPosC l2mCi24amP24g2 |amDg i 2 4oPa2

Pag2919

05

Leveraging this information and solution,

the

US Army

can have greater

Dgmt2Poygl2mCi2oycDs.g2

sD2dDgmPgD2yo||osC2|rffg||A2 2

OsD2Dgm|sC|2

isfrygCPgi2 oC2 Pao|2 cmcgD l2nsrD2Go|fs2 ccFnCmyof|2Pgmy2LsDymttn2Dg0rg|P|2 |csC|sD|aoc2mCi 2 gCdmdgygCP2LsD2PsoCP2ig.gtscygCP2hgP4ggC2Go|fs2 ccFnCmyof|l2QRM5GSvl2 QJ 2

Jo||osC2GsyymCi

A22

UsD3oCd

2

PsdgPagD

l 2

4g2fmC2

gCmhtg23gn2yo||osC2fmcmhot oPog|2LsD2dDgmPgD2mcctofmPosC2|rffg||2LsD2Pag2ep2 DynA2vag|g2

Cg42fmcmhotoPog|24ott2

Dgirfg2

ig.gtscygCP2mCi2scgDmPoCd2fs|P|2LsD2Pag2 Dynl2iDo.g2gLLofogCfog|2LsD2 05

Airborne

Division

and other combatant commands that employ MCIS applications. Our approach leverages commercial-off- the shelf capabilities to create a scalable and extensible platform currently in use across DoD and the Intelligence Community (IC) agencies. The platform provides the IT enterprise and application/MCIS developers the ability to gain situational awareness and gather real time actionable information that can be quickly and easily shared with all st akeholders for significantly reduced mean time to resolution.

The goal o

f this paper is to highlight the offering by AppDynamics, the work done, and benefits found in a field mission with the 101 05 , and to obtain sponsorship and engagement f or a joint development between Cisco AppDynamics, PEO-C3T, PM Mission Command and the NCFT. We would like to work together to enable these mission capabilities for greater application mission success for the US Army AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

Introduction

The 101

05

Airborne

Division requested our assistance to help identify and troubleshoot frequent deployment challenges with their Mission Command Information Systems (MCIS). Currently, the 101 05 uses basic troubleshooting techniques on the systems, applications, and networks that support their tactical applications. The systems are not as stable, robust or reliable as desired. Hours and days of time are spent resetting, reconfiguring, reimaging, an d replacing the components of the MCIS . This occurs over and over again during ex ercises and real world operations.

More often than not,

the root cause of the issues often experienced when operating MCI

S go undiscovered.

The result is a lack of

reliab ility and stability in the MCI S producing a mission outcome that is less than desired for the command. 2 ML2Pag2ymCn2fmcmhotoPog|2LsD2mcctofmPosC2cgDLsDymCfg2PamP2 ccFnCmyof|2fmC2igto.gDl2Pag|g24gDg2

Pag2PaDgg2mDgm|2sL2Lsfr|24oPa2Pag2919

05 during our recent exercise 9L

Can the 101

05 get the much needed but missing visibility into mission and command wide mcctofmPosC|l24aofa2

Pag2yo||osC2Dgmio

C g |2sL2Pag2fsyymCi A 2 2 IL What can be done to reduce the repetitive time and effort spent troubleshooting mission mcctofmPosC|l24agDg2 d gPPoCd2Ps2Pag2DssP2fmr|g2o|2DmDg A 2 SL What can be done to help baseline the mission applications when performing normally, |s24agC2cgDLsDymCfg2ig.omPg |2s.gD2Poyg2

Pag2919|P

2 fmC23Cs424oPa2fgDPmoCPn2Pag2Dgm|sC|2 4an A 2 pgg2LodrDg292hgts4A 2 DE,mrn, FmG;HmeImeJamfph,mp mfK,mIp,-FmIpLp MmM-pftKwmrwrf,hrm prrJ,rmfKufmNap MmFeO mfK,mhprrpe munn-ptufpe rPmu wmK,-nmK,a,m

OeJ-FmN,mfaJ-wmQu-JuN-,mfemJrlR

AprovedfrlasedfwrpithnDwrmyme.

AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved. 2

ApDynama

i a m T a cAyweygn a vag2DgymoCigD2sL2Pag2cmcgD24ott2cDs.oig2mC2s.gD.og42sL2 ccFnCmyof|l2|amDg23gn2LoCioCd|2LDsy2 srD24sD324oPa2Pag2919 05 , and request a plan of action to help resolve key pain points. 2 22
AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

Overview

Cisco's AppDynamics allows organizations to

provide optimal mission outcomes by improving application performance and user experience. It does this by providing a platform designed to follow the users or soldier's interaction with the application from begi nning to end.

AppDynamics allows real

time analysis and visualization of collected and correlated data to get insigh ts into IT operations, user experience and mission outcomesu Using machine learning

AppDynamics

automatically discovers all the business transactions that comprise the application and baselines (understands what normal is for) all metrics collected , so wh en a system has a slow or problematic application experience, developers can see exactly what caused the bottleneck and identify the issue quickly . This reduces the mean time to respond to issues, on average a 70%

90% improvement. AppDynamics

addresses the challenges brought by the t raditional approach of monitoring, where disparate teams try to solve application p roblems by using siloed IT infrastructure data to narrow down the root cause of a problem, lacking a holistic view.

AppDynamics Application

Intelligence provides the

Visibility to see

the Insight to k now and the recommending

Action to take. Specifically, that means:

o

8yey,ytydb: Ability to see the entire application (or set of applications) in real-time as to

individual end user experience, including t he supportive IT infrastructure; o

9ley:pd: Ability to know trends in the application over time, as well as its contribution to overall mission effectiveness; and

o

T;dynl: so, you can quickly fix problems or fine-tune an application in development based on data, reflecting deep visibility into how the app is relating with its surroun

ding infrastructure. AppDynamics adopts a top down approach to performance monitoring by uniq uely focusing on the end user's experience with something we call the Rick NcconaC cCsAkp c of the application. A Business Transaction traces the entire transaction path, from the end-user device or browser, through the application code, databases, third-party API calls, servers and other infrastructure. This i s a more accurat e way of understanding the health of applications than traditional botto m-

A"#"$%&'"

AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved. rc2mccDsmfag|2PamP2Dgtn2sC2|ots 5 hm|gi2ysCoPsDoCd2|strPosC|A22 MrD 2 mccDsmfa2o|2 ysDg2 .mtrmhtg2

24oPa2|otsgi2Psst|2mCi2

LoCdgD2csoCPoCdl2 ccFnCmyof|2hDoCd|2g.gDnsCg2

PsdgPagD2

LDsy2mtt2mcctofmPosC|l2|n|Pgy|2mCi2

fsycsCgCP|l2Ps2sCg2 asto|Pof2|strPosC2 aodatodaP oCd 2 mDfaoPgfPrDg2mCi2fsCPoCrg 2 2 |2 ccFnCmyof|2o|2 911Z2
|sLP4mDg2 hm|gil2oP2fmC2 nsrD2mcctofmPosC|2mDg2DrCCoCdA22vag2 ccFnCmyof|2mdgCP|2mDg2Pncofmttn2igctsngi2Ps2Pag2 tgD2ctmPLsDy2PamP2fmC2

Dg|oig2sC2

cDgyo|g2sD2oC2m2cDo.mPg2sD2crhtof2ftsri2gC.oDsCygCPl2|rfa2m|2 Up2sD2 6rDg2FMF2 ftsri A22

MrD2|strPosC2g©ctsoP|2Pag2t

K9Z 5

IZL2Ps2Pag2mcctofmPosC

A22 MCfg2igctsngi2 ccFnCmyof|24ott2io|fs.gDl2Pmdl2PDmfgl2fsDDgtmPg2mCi2 22
2 2

ApDynaFa

C a &RouslnabAopyouaeqacssxdgouAtra le: C uosarOe:AgpaosslAtoIAegaOnolIOaogba wornlAgnr a 2 2 2 2 2 2 22
tapiOlg9i: ;9nio,ni-f-ed

B"#"$%&'"

AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved. vag2 ccFnCmyof|2r|gD2oCPgDLmfg2sL 5 mcctofmPosC2 2 2

Global Applications View

Application Flow Map

Business Transactions

Tiers and Nodes

Service Endpoints

Database

Java and .NET Infrastructure metrics

Java and .NET transaction metrics

An example of such an "executive" dashboard is captured below in Figure 3. 2 2

Optimze@e

u e Apdixcwz,e,c-aghcm,Be-rznp=pneCDEF-eco,e=hmdcwwpotecmze llGeni-whd pHcgxze=hmece

4 ogeIia gcdesveJ6eDatdet ff sCeiCoecgmlC minengiogafl yePfgemifgf

C"#"$%&'"

AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

Initial Findings

dudmtL R

The AppDynamics platform was installed at the 101

05

Airborne and was ready to go within a day.

AppDynamics agents took longer to instrument the applications due to a n umber of factors b eyond immediate control, caused by incompatibility of MCIS with the program unless integrated at the development level. Once agents were temporarily hosted on three initial systems (See Figure 4), using AppDynamics, we immediately began collecting all observable metrics of the application code and the systems they were on. Within a few minutes we had some initial findings. R

AproveRe

Within a few hours we started to see (see Figure 5) data related to appl ication operation. We found on one server there was a health alert, so we drilled down into the alert to find that CPU was at 100% for several periodic intervals. We used AppDynamics to see all metrics from the virtual machine supporting the application. We can narrow the view for a certain time range to see if this KPI health alert is relevant to anything else observed.

D"#"$%&'"

AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

2 ApDynaSa

We also observed a node health alert, a memory race condition with one o f the JVM's, (see figure 6 . In this case the memory/HEAP usage alert lasted for hours, contribut ing to systems losing or dropping connection to each other, which had nothing to do with the network or the virtual machine. With our automated baselining, the system dynamically sets health rules to learn what is normal, and only alerts based on deviations from normal.

The value is that we

eliminate false positives such that IT operations can be focused on what matters, and not be distracted wading through a mountain of white noise false positive alert s. We also eliminate significant project costs because implementation teams do not require we eks or months of onsite consulting time to be setting and resetting thresholds to attempt to avo id this mountain of false positive alerts. E; AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

2 ApDynaTa

During this time frame we also collected and observed metrics on the dat abase. We see all database KPI's; functions, time spent, memory used, disk IO, network IO, and related performance metrics (see figure 7). A message from another system is being received but it is taking a long time. Again, this is unrelated to the network or the physical system running the application, but we are monitoring everything, tagging, tracing and corr elating events. With Project Manager (PM) integration before system fielding, Soldiers can b e provided pre- configured "watch lists" to identify proper/improper functioning. EE AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

2 ApDynaUa

Using the AppDynamics

dashboard, we click into the slow database method call to gain further detail (s ee figure 8 . Here we are presented the machine learned data as to the nature of th e JDBC call, the call statement itself, and related parameters that contributed to the slow message exchange, part of the slow business transaction that was measured. This level of detail is valuable for developers so that these findings can be used immediately i n an AGILE DevSecOps cycle, so that next release s of the application improve application performance and quality, which leads to the mission effectiveness goals that LTC Wilson and team are looking for at the 101
05

Airborne

and across the Army E AppDynamics Confidential. Copyright © 2020 AppDynamics LLC. All rights reserved.

2 ApDynaVa

All mission application transactions were m

easured from end to end, and the data can be sorted a number of ways, useful to trending and troubleshooting. We stack rank transactions based on which ones are slowest and

Java exceptions causing the condition

. We help see top transactions and load on t he systems. We see all the systems that participated in each transacti on, which ones were slow, and the root cause of why the transaction was slow (see figure 9). Just like the slow database interaction in figure 8, with the code level visibility we gather users have supreme operational dominance for your tactical applications, which ones are hav ing issues, and the rootquotesdbs_dbs17.pdfusesText_23

[PDF] mississippi goddam

[PDF] mississippi grand jury docket

[PDF] mississippi rules appellate procedure

[PDF] mississippi substitution of counsel

[PDF] missouri arraignment process

[PDF] missouri court docket records

[PDF] missouri court dockets search

[PDF] missouri probate court records

[PDF] missouri rules of civil procedure

[PDF] missouri shooting

[PDF] missouri state court docket

[PDF] missouri state court docket search

[PDF] missouri supreme court docket

[PDF] misterios develados saint germain pdf

[PDF] mit app inventor