In today's economy, IT departments are challenged by the differences between deploying distributed infrastructure and centralized infrastructure. Distributed infrastructure is appealing from an employee productivity and performance perspective, but it requires significantly higher capital expenditures (CapEx) and operating expenses (OpEx). Centralized infrastructure allows IT departments to consolidate application services into one or more data centers, thereby reducing CapEx and OpEx, but it presents performance challenges for remote employees who are held to the same productivity standards as employees closer to the application infrastructure.
Application performance management (APM) is a series of technologies that, when combined, enable managed services providers (MSPs), as well as IT departments that operate in a similar capacity, to understand how their applications are performing, provision the network according to business and application requirements, and optimize applications to help ensure consistent performance. Cisco® and Fluke Networks together provide an industry-leading solution for APM to enable MSPs to more intelligently manage performance of business-critical applications and enhance the investment in the network and application fabric that powers today's forward-looking businesses.
Application Performance Management
APM is a framework that takes advantage of visibility into application performance and network use to better align networking resources with business and application requirements and improve the performance of applications throughout the network. The benefits of an APM framework are clear: improved network utilization for business-critical applications, better performance for every user regardless of location, and understanding of network performance and utilization dynamics to assist in troubleshooting and enforcing performance service-level agreements (SLAs).
APM consists of three main components, each interoperating with the others to effectively accomplish the task of managing application performance (Figure 1):
• Visibility: By taking advantage of network data sources to understand what applications are on the network, how they operate, and how they are performing, IT departments can better adjust network configuration to improve utilization of precious network resources.
• Control: With the appropriate visibility, network resources can be provisioned through techniques such as quality of service (QoS) and performance routing (PfR) to help ensure that resources are used according to application performance requirements and business objectives.
• Optimization: With an understanding of how applications are performing, optimization techniques such as those provided by WAN optimization and application acceleration can be employed to reduce bandwidth consumption, enable consolidation of infrastructure, overcome the limitations created by latency and loss, and improve application performance for remote users.
Together, Cisco and Fluke Networks have developed an ecosystem of products that provide visibility, control, and optimization to manage end-to-end application performance.
Components of Cisco and Fluke Networks APM Solution
The Cisco and Fluke Networks APM solution consists of the following components (Figure 2):
• Fluke Networks Visual Performance Manager (VPM): This unified system provides integrated views and unique data correlation across multiple data sources to provide the visibility necessary for effective management of network, application, and voice-over-IP (VoIP) performance. Supported data sources include the Fluke Networks Application Performance Appliance, NetFlow Tracker, and Analysis Service Elements. Depending on the level of visibility required, one or all of these data sources can be integrated in Fluke Networks VPM.
• Fluke Networks Application Performance Appliance: This hardened, purpose-built data acquisition platform integrates with Fluke Networks VPM to provide actionable visibility of end-user response times for critical business applications, allowing the organization to optimize the delivery of essential business services. Deployed in the data center, Fluke Networks Application Performance Appliance can manage multi-tier application performance and report on actual service delivery with visibility down to the level of individual users and transactions.
• Fluke Networks NetFlow Tracker: This software-only or appliance-based solution uses NetFlow data collected from routers, switches, and other devices to provide insight into the effect of traffic on network performance. Collecting and keeping "all the flows, all the time," Fluke Networks NetFlow Tracker provides unique visibility into what is happening on the network and provides the data needed to make business decisions.
• Fluke Networks Analysis Service Element (ASE): LAN- or WAN-based probes provide 7-Layer visibility into application network, application, and VoIP performance from a networkwide point of view. Integrated into Fluke Networks VPM's scalable, multi-tenant architecture, Fluke Networks ASE provides a clear point of demarcation that users can use to determine whether a problem is caused by the application, network, or service provider.
• Cisco IOS®NetFlow: This component provides a critical set of services for IP applications, including network traffic accounting, network planning, security, and network monitoring. Cisco IOS NetFlow provides valuable information about network users and applications, peak usage times, and traffic routing.
• Cisco IOS QoS: This component provides network-integrated classification and resource provisioning capabilities that enforce priority and performance requirements within the network based on business and application needs.
• Cisco IOS PfR: This component provides network-integrated path selection based on application performance requirements and network performance metrics.
• Cisco Wide Area Application Services (WAAS): Cisco WAAS is an appliance-based and network-integrated system for overcoming performance limitations created by bandwidth limitations, inefficient application protocols, network latency, and packet loss.
Together, these components allow MSPs, and IT departments that operate similar to MSPs, to effectively use resources to improve end-to-end application performance, align network resources with business objectives, understand how applications operate on their networks, and quickly address performance problems when they arise.
Figure 2. Main Components of Cisco and Fluke Networks APM
Main Benefits of Cisco and Fluke Networks APM Solution
Table 1 summarizes the benefits of deploying a Cisco and Fluke Networks solution for APM.
Table 1. Main Benefits
Benefit
Description
Accurately baseline network and application performance
Using Fluke Networks Application Performance Appliance to passively measure network traffic, Fluke Networks NetFlow Tracker (collecting and mediating NetFlow), and Fluke Networks ASE (monitoring WAN traffic and VoIP), organizations can use the Fluke Networks VPM to quickly and accurately understand how the network is being used, how applications are performing, VoIP call quality, and where application performance bottlenecks may exist. Additionally performance baselines can be established, and thresholds can be set to trigger alarm conditions when deviation from the baseline is detected.
Determine which applications are good candidates for acceleration
Application performance metrics are visualized in detail in Fluke Networks VPM, highlighting data volume and rate, response-time metrics, and other performance-related statistics. This data helps identify applications that are experiencing performance challenges, including throughput limitations, high response times, and high levels of packet loss, which can indicate that the application or site is a good candidate for deployment of Cisco WAAS to improve application performance.
Apply acceleration to improve application performance and validate benefits against baseline
Cisco WAAS provides WAN optimization and application acceleration techniques to reduce the effect of factors that limit application performance, including bandwidth constraints, high latency, high packet loss, and chatty protocol behavior. By continuously monitoring application performance through Fluke Networks VPM, customers can monitor and report the performance improvement provided by Cisco WAAS, validating the benefits and quantifying the improvement and effect on the business.
Intelligently utilize available network paths according to application needs
By taking advantage of the application performance visibility and network baselining capabilities provided by Fluke Networks, organizations can use Cisco IOS PfR to align application traffic with the appropriate network paths to enhance application performance and increase efficiency.
Help ensure alignment of network resources with application performance requirements
With an understanding of how applications are using the network, organizations can employ end-to-end QoS from within Cisco IOS Software to align network resources with the performance requirements demanded by the business as well as the applications themselves.
Consolidate costly remote-office infrastructure
With the network provisioned according to business priority and application requirements, along with acceleration techniques for improving application performance, organizations can confidently begin consolidating costly remote-office infrastructure into centrally managed data centers or MSP hosting facilities. Doing so dramatically reduces ongoing CapEx and OpEx.
Effectively Baseline Your Network and Application Performance
To proactively manage performance, you need to characterize what "normal" performance is (Figure 3). With this information, you can quickly recognize and provide notification when degraded performance is occurring and determine the severity of the degradation; which applications, sites, or end users are affected; and the likely root cause.
Figure 3. What Is Normal Performance?
Determine Which Applications and Sites Will Benefit from Optimization
By analyzing critical performance metrics and response times experienced by end users for business-critical and customer-facing applications, Fluke Networks VPM provides the data that is needed to understand which applications and sites are the best candidates for WAN optimization and application acceleration (Figure 4). This analysis provides a baseline from which the improvements provided by Cisco WAAS can be clearly quantified and visualized.
Figure 4. Which Sites and Applications Are Performing Poorly?
Apply Optimization and Quantify Performance Improvement
With an understanding of which sites and applications are good candidates for optimization, Cisco WAAS can be deployed in data center and branch-office locations to reduce the negative effect of WAN conditions on the performance of your applications. Cisco WAAS employs powerful WAN optimization techniques (that are application agnostic, providing benefits to essentially any TCP-based application) as well as application acceleration techniques (that are application specific, helping overcome the challenges provided by specific applications and protocols) to overcome WAN conditions and help ensure consistent high-performance access to applications and information for essentially any corporate user.
Cisco WAAS provides optimization for the applications that are critical to your business, and in many cases the improvement is five times or greater (Figure 5).
Figure 5. Cisco WAAS Accelerates the Applications You Care About
With Cisco WAAS deployed with Fluke Networks VPM, customers can see "before and after" data showing the performance improvement provided by Cisco WAAS. Having the capability to quantify the performance improvement is critical as it provides organizations with tangible data proving the improvement and enables genuine calculation of return on investment (ROI). Figures 6 and 7 show how Fluke Networks VPM compares the performance of an application before and after Cisco WAAS implementation and measures the effectiveness of various Cisco WAAS optimization and acceleration techniques.
Intelligently Use Network Paths According to Application Needs
Cisco IOS PfR provides a natural complement to Cisco WAAS. Cisco WAAS optimizes TCP sessions, while Cisco IOS PfR can automatically determine which network path has the performance characteristics most closely aligned with various types of traffic. Working in concert, these two technologies deliver optimized sessions over optimal network paths (Figure 8).
Figure 8. Cisco IOS PfR and Fluke Networks PfR Manager
Align Network Resources with Business Priorities
With Fluke Networks VPM in place for full visibility into application performance and Cisco WAAS to optimize application performance over the network, organizations can take advantage of the QoS capabilities of Cisco IOS Software to align network resources according to the requirements of applications or the corresponding business priority. Cisco IOS QoS provides a foundation that allows a network to handle the multitude of applications and data while helping ensure secure, predictable, measurable, and sometimes guaranteed services to specific applications. QoS helps ensure that your network provides the appropriate level of service for your applications according to the need of the application and the importance of the application to the company's objectives. Capabilities provided by QoS include:
• Classification: Applications are identified accurately based on numerous detection techniques, including advanced methods that take into consideration the actual content traversing the network.
• Marking: Differentiated Services Code Point (DSCP) markings are applied to packets according to classification to help ensure consistent handling throughout the customer and MSP networks.
• Intelligent servicing: Using numerous queuing and scheduling techniques, traffic can be serviced according to priority or performance requirements.
• Bandwidth control: Policing and shaping control bandwidth utilization to provide critical applications with the resources and handling they need, while also helping ensure that recreational or scavenger traffic does not affect applications critical to your business.
Centralize and Consolidate Costly Remote-Office Infrastructure
One element that contributes most dramatically to the ROI enabled by the Fluke Networks and Cisco end-to-end APM solution is the capability to confidently consolidate distributed infrastructure from remote offices into one or more managed data centers. By providing application acceleration and WAN optimization capabilities, Cisco WAAS provides remote users with levels of performance similar to those they are accustomed to with local resources deployed onsite. Cisco WAAS provides the capability to move the following devices and services from remote offices and consolidate them within data center locations, where utilization and manageability efficiencies further improve ROI:
• Server components: Server hardware can be centralized in the data center, and the number of servers may be reduced through the intelligent acceleration provided by Cisco WAAS. Cost reduction includes server hardware, operating system licenses, and patch management.
• Application components: With consolidation of the server infrastructure, applications can be deployed centrally rather than in a distributed manner, while providing similar levels of performance over the WAN.
• Storage components: Data protection components and external storage can be centralized in the data center onto more scalable systems that provide lower cost of operation per megabyte or gigabyte, and the number of elements associated with data protection in the branch office can be reduced.
• Ongoing administration: By consolidating remote-office infrastructure in the data center, ongoing costs associated with IT infrastructure management in each remote office are reduced, thereby providing substantial savings when this model is employed throughout the organization.
For those services that cannot be consolidated, Cisco WAAS virtual blades can be employed to locally host infrastructure services such as those that run on Microsoft Windows Server 2008, including read-only domain controller (RODC), Domain Name System (DNS) services, Dynamic Host Configuration Protocol (DHCP) services, and print services.
As well as enabling centralization of infrastructure, Cisco WAAS enables consolidation of file server and video server resources that have been centralized, reducing the number of servers necessary to support the global workforce. Application acceleration provided by Cisco WAAS provides intelligent protocol management, data caching, and metadata caching. These features allow the network as the platform to begin offloading workload from your data center servers when safe to do so. The result for your application infrastructure is a dramatic reduction in server workload, thereby allowing you to reduce the number of data center file and video servers necessary to support a global workforce.
Figure 9 shows how Cisco WAAS safely offloads data center file servers and network-attached storage (NAS) devices, and Figure 10 shows how Cisco WAAS safely offloads data center video servers.
Figure 9. Cisco WAAS Safely Offloads Data Center File Servers and NAS Devices
Figure 10. Cisco WAAS Safely Offloads Data Center Video Servers
Conclusion
By taking advantage of APM solutions from Cisco and Fluke Networks, MSPs and IT departments that operate in a similar manner have full visibility into application performance and can align network resources and control utilization according to business priority and application requirements and accelerate the performance of applications to essentially any user. These technologies provide the foundation for visibility, control, and optimization and enable the organization to employ SLAs based on application performance metrics to help ensure consistent performance throughout the network and provide the information necessary to confidently analyze essentially any performance challenge that may arise in the future.
