Slow Applications and Sluggish Networks Slow Revenue

Slow applications impact businesses more than we think. Many think that application performance only affects users accessing information and productivity of employees. However, a study shows that poor application performance has a direct impact on revenue.

Aberdeen Research surveyed 200 organizations to learn about the impact of poor application performance. About 60% reported that they were not happy with performance of applications that were critical for their business. They also said that slow applications impacted their corporate revenues by up to 9%. Other common problems of poor application performance were:

• 58% respondents experienced decline in employee satisfaction
• 47% respondents saw decline in responsiveness to the needs of users
• 32% respondents mentioned damage to their brand reputation
• 31% respondents reported reduced productivity of their staff

There are many reasons that make maintaining application performance difficult. The most important one is inability to identify issues before end users are impacted. Poor network documentation often causes this delay. Up to date network diagrams are critical for real time troubleshooting. 45% of the surveyed enterprises were concerned about the lack of performance visibility as they continue to add more business applications. The problem with application performance can be summarized as lack of visibility into network and performance of applications.

The four step Map-Driven PHD methodology makes uncovering these issues exponentially faster. This network troubleshooting method leverages advanced visualization and automation to minimize the use of the CLI.

Step 1: Map the Application Path Instantly

You can instantly create a layer-2 or layer-3 path of the application flow with automation that leverages live gateway information and routing tables.

Step 2: Probe Performance Hotspots

With multiple devices in the path, trying to find which link is congested or dropping packets is difficult. You can turn on live monitoring with one click to view link status, memory utilization, packet loss and more.

Step 3: Historical Analysis: Compare Network Status against a Baseline

Whether you suspect a routing, configuration, or topology change is to blame, you can instantly uncover what’s changed to cause the problem. Automatically compare route tables, configuration files, and show-command output against a working baseline to highlight what’s changed. You can also draw the application path using historical data so you can visualize what’s changed.

Step 4: Drill-Down for Root Cause

If you suspect congestion, use NetFlow to drill-down top-talkers. For hundreds of other causes, leverage automation procedures to find the root cause. Helpful procedures include ‘CPU Overutilization’, ‘Queue Drop’, ‘Link Congestion’, and ‘Flapping Links’.

Businesses need to understand their network better to maintain healthy revenues, enhance staff productivity and user experience. What application problems affect you the most? Do you have other, more efficient ways of maintaining application performance? Let me know by commenting below.

Seven Rules to Avoid/Shorten Network Outages

Apple’s service webpage said on Tuesday that “some” users were affected by the five hour outage, and also revealed that FaceTime was down for some users too. Not many people were seen to be complaining about FaceTime but the same could not be said about iMessage. Text messaging is a backup but not for those who use iPads or iPod touch without data plans.

Source: Apple Services, Stores, and iCloud

Tuesday’s Twitter rage suggests that a lot of users were upset because of this outage. A user said, “My iMessage is down… Now I’m stuck sending normal text messages like a common peasant,” while hundreds of others speculated that the outage was due to an attack from North Korea.

Apple did not comment on the cause and scale of this outage. Apple Insider reports that this is the fourth bit of downtime to plague iMessage over, “the past three months.” Many tech bloggers continue to use this outage to pick on Apple because of all their recent problems. However, you will be surprised to know how common such outages are.

Billions Are Lost Each Year Due to IT Outages

A CDW survey of 7,000 businesses in 2010 reveal:

• 25% of enterprises suffered unplanned outages of 4 hours or more
• These outages cost enterprises $1.7 billion in lost profits
• About ¼ of outages are network-related

A separate Cisco study uncovers the causes of network outages:

• Router or Switch failure (23% by software/hardware and DOS attack)
• Link failure (32% by fiber cuts, network congestion )
• Network Change (36% by upgrade and configuration change)
• Misc. (9%)

Survey results challenge our assumptions of being protected by redundant routers and switches, fault-tolerant circuits and change management processes that review every change. Many think that “doubling-down” is the answer but it is certainly not enough.

Seven Ground Rules to Avoid or Shorten Network Outages

• Baseline Your Network Religiously
• Update Your Documentation Often
• Insist on True Network Visualization Capability
• Test Every Change Thoroughly
• Look at Fault-tolerant Design Carefully
• Document Your Troubleshooting Process
• Invest Time in Cross Training

In addition to existing protection, enterprises need to apply automation to defend their network. Human errors must be prevented during design and implementation.  Troubleshooting should not take days; user’s voice would have been much louder had iMessage been down for 2-3 days. There should be sufficient collaboration and training among teams in order to react quickly in such situations.

Where Are All the Multicast Experts When You Need Them?

Seasoned network engineers wear many hats and are fluent in complex technologies from routing to spanning-tree. But why are so few professionals comfortable with multicasting when it is so prevalent in today’s network communications? Simply Hired reports that multicasting job postings have increased by 63% in the last 18 months. From my observations, the number of qualified multicast specialists isn’t satisfying this demand.

Multicast Job Trends

It’s clear why multicasting demand is growing. The financial industry is reliant on multicasting for distributing market data and stock updates. Enterprise video and collaboration are on the rise with technologies like Cisco TelePresence, DMB, and WebEx all increasing in popularity. The entertainment industry is growing more reliant on multicasting for cable services and Video on Demand.  In fact, here is a list of popular applications that use multicast.

Popular Multicast Applications

With such an increased demand for multicasting, why aren’t there more multicast specialists? I think it’s because multicasting technology is just too complicated. Multicast forwarding is dynamic in nature and difficult to visualize. The terminology is complex and the underlying concepts are elaborate. What’s the difference between a Shared Tree (*, G) and (S, G)? How is a Designated Router elected? What if the DR goes down? How are Shared Trees Pruned to stop the RP from forwarding traffic to a downstream router?

Shared Tree Pruning

During troubleshooting, engineers often need to know if a PIM interface or PIM neighbor has changed. Most importantly, they need a way to visualize the Multicast Distribution Tree (Shared Tree or Source Tree) and map the First Hop Router, Last Hop Router, Reporters etc. With such high value information as financial data on the line, imagine the pressure on the engineer when multicast problems occur. The right tool can help lower the barrier to understanding, visualizing, and troubleshooting multicasting.

Automatically Detect PIM Neighbor Changes

Automatically Map MDT

If engineers could more easily visualize the flow of multicast traffic down the source tree and understand dynamic changes in multicast configuration, I think more of them would consider themselves ‘multicast experts’.

Configuring DHCP Relay Agent | Complete Lab Included

A DHCP relay agent is any host that forwards DHCP packets between clients and servers. Relay agents are used to forward requests and replies between clients and servers that are not on the same physical subnet.

Initially, R1 has been configured as a DHCP client as shown below. R3 was configured as the DHCP server for network 11.11.12.0/24. In this lab, we configure DHCP relay Agent on R2 to ensure that DHCP’s broadcast packets can cross R2 to R3.

 

Network Diagram: Configuring DHCP Relay Agent

NetBrain’s unique lab format (Qmap) gives you access to step-by-step DHCP Relay Agent configuration and show-commands which highlight the impact of the configuration. You will need NetBrain Qmap Reader to view these details with an in-depth lab topology diagram.

Download Qmap

Note: If you don’t have NetBrain Qmap Reader already, click here to download it. To view more labs, visit Qmap Learning Center.

Lab Credit: Frank Tan

Spanning Tree Root Bridge | Complete Lab Included

With Spanning Tree Protocol (STP), it is important for all the switches in the network to select a root bridge which is then the focal point in the network. Other decisions such as which port to block and which port to put in forwarding mode, are made from the perspective of this root bridge.

Each VLAN must have its own root bridge because each VLAN is a separate broadcast domain. The roots for the different VLANs can all reside in a single switch or in various switches.  The selection of the root switch for a particular VLAN is very important. You can choose the root switch, or you can let the switches decide, which is risky.

This lab shows how to configure the most capable switch as the root bridge. Initially SW3 is the root bridge in this spanning-tree domain. Some of SW2 ports are blocked by spanning-tree. The objective is to ensure SW2 is the Root Bridge in the spanning tree domain so it does not have any ports blocked by the spanning tree.

Network Diagram: Configuring Spanning Tree Root Bridge

NetBrain’s unique lab format (Qmap) gives you access to step-by-step Spanning Tree Root Bridge configuration and show-commands which highlight the impact of the configuration. You will need NetBrain Qmap Reader to view these details with an in-depth lab topology diagram.

Download Qmap

Note: If you don’t have NetBrain Qmap Reader already, click here to download it. To view more labs, visit Qmap Learning Center.

Lab Credit: Frank Tan

Using VTP to Simplify VLAN Configuration | Complete lab Included

VLAN Trunk Protocol (VTP) reduces administration in a switched network. When you configure a new VLAN on one VTP server, the VLAN is distributed through all switches in the domain. This reduces the need to configure the same VLAN everywhere. VTP is a Cisco-proprietary protocol that is available on most of the Cisco Catalyst series products.

In the lab, we’re configuring VTP between SW1 and SW2 with domain “NetBrain” and password ‘cisco’. SW1 should be the Server and SW2 should be a Client. After creating VLANs on the VTP, server SW1 and SW2 should both be able to learn all VLANs.

Network Diagram: VTP for VLAN Configuration

 

NetBrain’s unique lab format (Qmap) gives you access to step-by-step configuration and show-commands which highlight the impact of the configuration. You will need NetBrain Qmap Reader to view these details with an in-depth lab topology diagram.

Download Qmap

Note: If you don’t have NetBrain Qmap Reader already, click here to download it. To view more labs, visit Qmap Learning Center.

Lab Credit: Frank Tan

EIGRP Route Filtering by Metrics | Complete Lab Included

In EIGRP, metric values can be manually changed for interfaces.  Once metric values are set, routes whose metrics fall within a certain range can be filtered with a route-map.

EIGRP is configured on R1, R2, and R3 with default metric 409600 at the start of the lab. Our objective is to manually set metrics for R3’s and R2’s lo192 interfaces. In the next step, we configured a route map to filter out traffic to R2’s and R1’s lo0 interfaces using metrics.

Network Diagram: EIGRP Route Filtering

NetBrain’s unique lab format (Qmap) gives you access to step-by-step EIGRP Route Filtering configuration and show-commands which highlight the impact of the configuration. You will need NetBrain Qmap Reader to view these details with an in-depth lab topology diagram.

Download Qmap

Note: If you don’t have NetBrain Qmap Reader already, click here to download it. To view more labs, visit Qmap Learning Center.

Download All EIGRP Qmaps

Lab Credit: Tim Girouard

EIGRP Passive Interface Configuration | Complete Lab Included

An EIGRP passive interface will not form EIGRP neighbor relationships or send EIGRP updates and thus will not be added to EIGRP route tables. This lab focuses on disabling R3 routing through R2 by configuring its connecting interface (f0/1) as a passive interface.

There are two ways to accomplish the lab objective:

• Configure all R3 interfaces as passive, and then activate the f0/0 interface
• Configure all R3 interfaces as active, and then make all interfaces except the f0/0 interface passive

Network Diagram: EIGRP Passive Interface

NetBrain’s unique lab format (Qmap) gives you access to step-by-step EIGRP Passive Interface configuration and show-commands which highlight the impact of the configuration. You will need NetBrain Qmap Reader to view these details with an in-depth lab topology diagram.

Download Qmap

Note: If you don’t have NetBrain Qmap Reader already, click here to download it. To view more labs, visit Qmap Learning Center.

Download All EIGRP Qmaps

Lab Credit: Tim Girouard

BGP Routing: Propagating Default Routes | Complete Lab Included

Border Gateway Protocol (BGP) can create default routes in two ways: by injecting it into BGP with a static route, and by generating a conditional default route that is sent to a configured peer.  The former requires there to be a default route already in the RIB.

On BB1, we advertise a static default route to an external network. In the next step, we advertise a route with the condition that the fa0/0 interface is up.

Network Diagram: BGP Routing

NetBrain’s unique lab format (Qmap) gives you access to step-by-step BGP Routing configuration and show-commands which highlight the impact of the configuration. You will need NetBrain Qmap Reader to view these details with an in-depth lab topology diagram.

Download Lab

Note: If you don’t have NetBrain Qmap Reader already, click here to download it. To view more labs, visit Qmap Learning Center.

Lab Credit: Tim Girouard

EtherChannel Configuration for Cisco | Complete Lab Included

An EtherChannel is used to create higher bandwidth interfaces by combining ports. This lab demonstrates how to configure a layer-2 EtherChannel between switch SW1 and SW2 and a layer-3 EtherChannel between switches SW3 and SW4.

NetBrain’s unique lab format (Qmap) gives you access to step-by-step EtherChannel configuration and show-commands which highlight the impact of the configuration. You will need NetBrain Qmap Reader to view these details with an in-depth lab topology diagram.

Network Diagram: EtherChannel Configuration

 Technology Overview

1. A layer-2 EtherChannel is used between two switches and often configured as a trunk. Spanning-tree interprets the channel as a single link so both ports will share the same forwarding/blocking state.
2. A layer-3 EtherChannel configuration is done on a router or as a routed interface on a switch with a single IP address.

The choice of L2 vs L3 EtherChannel is similar to the decision of configuring a port as an L2 (switch port) or L3 (no switchport).

Download Lab

Note: If you don’t have NetBrain Qmap Reader already, click here to download it. To view more labs, visit Qmap Learning Center.

Lab Credit: Frank Tan