Improve Network Latency and Hold Service Providers to SLAs

ca Opscenter

Improve Network Latency and Hold Service Providers to SLAs

Ahmed Kira

OCX64S #CAWorld

CA TechnologiesService Assurance Sales Engineering

2 © 2014 CA. ALL RIGHTS RESERVED.

Abstract

A leading pharmaceutical company leverages CA Network Flow Analysis, CA Application Delivery Analysis and CA eHealth to improve network latency by up to 39% and hold their service providers to their Service Level Agreements.

Ahmed Kira

CA Technologies

Advisor,

Service Assurance Sales Engineering


Agenda

BACKGROUND

BEST PRACTICES & SOLUTION ARCHITECTURE

Q&A

USE CASE 1: REDUCING NETWORK LATENCY BY 39%

USE CASE 2: PROACTIVELY HOLDING SERVICE PROVIDER ACCOUNTABLE

USE CASE 3: FAULT ISOLATION

1

2

3

4

5

6

Background


Background

20,000 Employees

Leading Pharmaceutical Company

60+ Global Sites

Pioneer in treatingbone and kidney cancers

Innovative research with

‘biology first’

12,000 devices

6+ million netflow records/minute

11 Global Data CentersAlmost $20 billion Annual Revenue

Best Practices


IPSLA/Response monitoring across links– Edge router to router

– Eliminates LAN, which Service Provider is not responsible for

– Can emulate different Quality of Services

Network Connection Time– Based on packet analysis

– Calculated from real transactions, not synthetic; no period missed

– Not subject to ‘Delayed Acks’, which impact ‘Network Round Trip Time’

– Averaged across many transactions

What metrics needs to be tracked?

BEST PRACTICES


User Server

ApplicationTransaction

ADA Collector

NoResponse

Data Response B2

Server Connection Setup Time

Network Connection Setup Time

ACK Response B2

ACK Response B1

Data Response B1

NoResponse

Data Response B1

Data Response B1

HTTP GET index.html (Data Request B)

ACK Response A2

ACK Response A1

Data Response A1

HTTP GET index.html (Data Request A)

Data Response A2

TCP(80) SYNTCP(80) SYN ACK

TCP(80) ACK

Service ProviderNetwork


Network Round Trip Time & Network Connect Time

SerializationDelay

Queue Delay

ForwardingDelay

DistanceDelay

ProtocolDelayNRTT

Fiber:5.5us/km

Satellite:3.3us/km

MediaAccess

DelayedACKsPer Hop Behaviors

NCTSmaller frame

size means lowerserialization

delay

NoDelayed

ACKs


Network Round Trip Time & Network Connect Time

NCT

NRTT

Consistent

More suited for when analyzing

specificapplication traffic;

Tuning to excludekeep-aliveprocessing


CA Solution Architecture in production

CA Spectrum

CA Network Flow Analysis (NFA)

Network Traffic Flow Analysis

CA Performance Center

CA Application Delivery Analysis (ADA)

Integrated

Workflow

Centralized Fault ManagementCentralized Performance Management

Exceptions

3rd Party

Alarms

- L2-L3 Auto-Discovery- Topology Visualization- Network Configuration Management- Root Cause Analysis- Asset, Alarm, and Availability Reporting

CA eHealth

Abnormal Application Performance Detection Proactive Investigation

Device management IPSLA

Use Case 1:Decreasing network response by 39%


Background

Relatively high costs was original motivator

Need for additional capacity to accommodate growth

Lack of responsiveness and accountability from prior service provider

Unacceptable SLA performance for errors and out of order packets

New vendor selected to address business needs


Prior to circuit cutoverNetwork Connection Time Retransmission Delay

Network Connection Time average of 22 ms, while acceptable, not prime

Retransmission delay considered unacceptable


During Circuit Cutover

No more Netflow from old interface as it decommissioned around 13:25

39% drop in network

connect time!

Network Connection Time Retransmission Delay

Promising retransmission

delay


After Circuit CutoverNetwork Connection Time Retransmission Delay

Stable NCT at under 14 ms. Notice old baseline! A 39%

reduction!

Continued positive trend!

Negligible retransmission

delay!


Results

Changing service resulted in– reduced costs

– improved quality of service

Problems primarily documented through use of– CA Application Delivery Analysis (leveraging Network Connect Time

and Retransmission Delay for specific network location filters)

– CA Network Flow Analysis

Use Case 2:Proactively holding Service Provider accountable


Background

Multiple links with same network service provider between key data centers and locations

SLA includes traffic load balancing.

Benefits of load balancing:– No one circuit has heavy utilization, thus reducing network queuing

delay and contention

– Extra bandwidth available for busy periods

Traffic Load Balancing


Was any traffic load balanced?

Upon setting up views shortly after circuits provisioned, administrators

noticed no load balancing was occurring

Uneven flow of traffic

across paired circuits


ResultsHold service provider accountable

Detailed Insight

By leveraging CA Network Flow Analysis:

Evident that uneven load balancing was occurring

Provide additional analysis whether behavior was due to a type of service or protocol mapping

Proactive

Detected situation before it becomes a problem with heavier traffic throughput.

Accountability

Escalated to service provider to hold them accountable to their SLA.

Use Case 3: When the Service Provider is NOT at fault


Background

Overseas location in Europe complaining of slow application response times. Blaming the network!


Is it the server?

Leveraging, eHealth views in Performance Center,

customer validates low server load


Network Consistency

Distance between locations is 9,000 km (6,000 miles)

Filtered for non WAN-optimized traffic

Network Connection Time over 1 week is consistent at around150 ms. No significant spikes in network performance.

Retransmission delay negligible with few peaks to 1 ms.

Not thenetwork

!


ResultsSpeedy & Effective Fault Determination

ID Fault Domain

The issue was concluded to NOT be the network, so the provider was not to blame.

Real Root Cause

Using CA Application Delivery Analysis, it was determined that the application was making thousands of queries per minute.

Many TCP sessions

1 User

Actionable

Host the application closer to the database server

Or

Use database replication and access local database

Or

Tune applications to make fewer queries


Summary

Key topics

A leading global pharmaceutical company with over $100 Billion Market Capitalization needs to reduce cost of network services, yet improve quality of service for corporate and business partner end-users

Findings

By Leveraging CA Application Delivery Analysis (ADA), CA Network Flow Analysis (NFA), and CA eHealth, the pharmaceutical company proactively tracks service provider latency, network response times, and ensure service providers are fulfilling their SLA obligations.

Experiences

- Reduced network latency by up to 39%

- Proactively ensure optimum application response times

- Assure that Service Provider is within latency SLA


For More Information

To learn more about DevOps, please visit:

http://bit.ly/1wbjjqX

Insert appropriate screenshot and text overlayfrom following “More Info Graphics” slide here;

ensure it links to correct pageDevOps






For Informational Purposes Only

This presentation was based on current information and resource allocations as of August 2014 and is subject to change or withdrawal by CA at any time without notice. Not withstanding anything in this presentation to the contrary, this presentation shall not serve to (i) affect the rights and/or obligations of CA or its licensees under any existing or future written license agreement or services agreement relating to any CA software product; or (ii) amend any product documentation or specifications for any CA software product. The development, release and timing of any features or functionality described in this presentation remain at CA’s sole discretion. Notwithstanding anything in this presentation to the contrary, upon the general availability of any future CA product release referenced in this presentation, CA will make such release available (i) for sale to new licensees of such product; and (ii) to existing licensees of such product on a when and if-available basis as part of CA maintenance and support, and in the form of a regularly scheduled major product release. Such releases may be made available to current licensees of such product who are current subscribers to CA maintenance and support on a when and if-available basis. In the event of a conflict between the terms of this paragraph and any other information contained in this presentation, the terms of this paragraph shall govern.

Certain information in this presentation may outline CA’s general product direction. All information in this presentation is for your informational purposes only and may not be incorporated into any contract. CA assumes no responsibility for the accuracy or completeness of the information. To the extent permitted by applicable law, CA provides this presentation “as is” without warranty of any kind, including without limitation, any implied warranties or merchantability, fitness for a particular purpose, or non-infringement. In no event will CA be liable for any loss or damage, direct or indirect, from the use of this document, including, without limitation, lost profits, lost investment, business interruption, goodwill, or lost data, even if CA is expressly advised in advance of the possibility of such damages. CA confidential and proprietary. No unauthorized copying or distribution permitted.

Terms of this Presentation

Copyright © 2014 CA. All rights reserved. All trademarks, trade names, service marks and logos referenced herein belongto their respective companies. CA confidential and proprietary. No unauthorized copying or distribution permitted.

Technology

Improve Network Latency and Hold Service Providers to SLAs