Choosing the Right Internet Traffic and Content Management Solution
By David Barclay (Issue 3 2000)
Many products are available to help businesses enhance the performance, reliability, and economic expansion of their Internet site. These products all support a concept known as Internet Traffic and Content Management (iTCM). This article describes various iTCM solutions, including Dell's PowerApp.BIG-IP, to help you decide which one is right for your business.
The rapid growth of the Internet has caused an increase in the deployment of various Web technologies. It also has contributed to the overall complexity of managing content, networks, and applications. As businesses look to leverage the expanding Internet market, they face an ever-growing gamut of protocols to support, devices to manage, and content and applications to deploy.
In addition, Web traffic through certain sites doubles every 100 days,1 and network computing hardware often cannot keep up with the increased demands. For organizations such as Web hosting companies, dot-coms, and Internet service providers to remain competitive, they need a reliable solution for handling this traffic.
Various products are available to enhance the performance, 24x7 reliability, and economic expansion of public and private Internet sites. Solutions that offer these capabilities come in many varieties, but all serve to facilitate a concept known as Internet Traffic and Content Management (iTCM).
What distinguishes these products from one another? What makes one solution more viable than another? The current flood of messages from various vendors, analysts, and reviewers has done little to clarify the trade-offs. Rather, it has only added to the confusion.
Growing Demand for Internet Business Applications
While business use of the Internet began conservatively, it has rapidly grown into a sophisticated array of e-commerce and content personalization applications for consumers and businesses alike. Figure 1 shows how the value of Internet business applications has increased in proportion to their complexity.
Figure 1. The Growth of Internet Business Applications
The bottom line is that businesses are extending their online focus to applications, networks, protocols, and content, which are becoming more complex each year. Combined with this shift is the need to provide fail-safe access to accurate content, while increasing processing capacity and availability, and simplifying management.
Internet Traffic and Content Management Defined
As Internet sites begin to handle more traffic and support more complex protocols and services, availability and fault tolerance become critical needs. Every transaction and user interaction must be 100 percent reliable to maintain optimal quality of service (QoS). To address these needs and prevent overload to one specific server, sites often replicate data across an array of servers.
As more servers are deployed, the site becomes more costly and more difficult to manage, and there is little assurance that one server will not become overloaded, provide incorrect responses, or outright fail. The site needs a more intelligent product that can complement and manage incoming traffic—a function known as load balancing . With load balancing, traffic can be dynamically distributed across a group of servers running a common application—yet the group appears as one server to the network.
Load balancing distributes traffic more efficiently, offers greater economies of scale, and provides significantly greater fault tolerance. Internet traffic and content management products, however, not only encompass load-balancing capabilities, they also intelligently monitor and manage the health of servers and Internet content, and make decisions on where to route traffic to optimize site performance and availability. This intelligence ensures that users are connected to the most available server, providing excellent and predictable QoS and the right content.
Preventing Service Interruptions and Guaranteeing Availability
With today's Web applications, service interruptions can be costly and can occur in many forms. Server and software failures are the most common, because hardware, operating systems, and applications may simply stop responding. Content failures, error messages, and incorrect data can infuriate users. Finally, heavy traffic and network congestion or failure can easily limit site availability.
The iTCM products, therefore, must be designed to prevent these interruptions and guarantee availability. A solution not geared to provide high availability does not maximize the return on investment for Internet and intranet connectivity. Therefore, as you evaluate products, you should look for a balance of the following:
- QoS-based availability
- Assured and continuous operation with zero downtime
- Simplified, consistent management across a range of protocols
- Robust technical support and ease of installation
Using a single solution to provide these critical elements ultimately can provide tremendous cost savings, enhance user experience, and provide significant long-term business value.
Internet Traffic and Content Management Solutions
iTCM products typically offer varying levels of expandability, availability, and speed. The following sections describe the various product offerings.
With software-only solutions, software is installed directly onto the servers in the array. This enables network managers to perform highly granular server management operations, such as analyzing CPU memory utilization and managing agent-based content. In theory, software-only solutions provide cost savings and faster performance because traffic does not have to pass through an additional device. In addition, loading software onto all servers in the array eliminates a single point of failure, since the entire site will not go down should one server fail.
Some software-only products allow synchronization of data among servers in a cluster. This feature can be useful if the servers do not maintain identical content and do perform complementary tasks, which require different servers to work in concert to complete a single content request.
Although software-only products eliminate a single point of failure and enable in-depth analysis and synchronization of data, there are trade-offs to recognize with this solution. Most notably is server management. Since the software is loaded on all servers in the array, maintaining this software can become task-heavy over time, since all servers will require updates.
Another consideration for software-only solutions is OS dependency. Because the software is installed directly onto the servers, businesses are locked into supporting specific platforms. Security concerns also can arise because this type of solution exposes server IP addresses directly to the user.
Switches can perform fast load balancing at layers 2 and 3 in hardware managed by a central processor, which executes background tasks such as routing, table, and network management. This type of solution enables fast balancing of static content, and many solutions offer high backplane-speed support. In addition, switches have the potential to connect to multiple high-speed Ethernet ports simultaneously, further optimizing speed. But one irony to this configuration is that while switches provide multi-gigabit throughput, most sites connect to the Internet using T3 (45 Mbps) or less lines, rendering the additional capability moot.
This switch-based architecture, however, has limitations. For example, packets that require exception handling at the network level (layer 4) must be separately opened and examined to determine their destination port. This process uses the switch's central processor and may compromise its performance. In other words, executing per-frame processing from a centralized processor can limit the total frame throughput of the device.
Often missing from switch solutions are functions such as Secure Sockets Layer (SSL) session ID tracking, user authentication, and application health checking. Their absence further limits the intelligence and viability for more sophisticated tasks, such as e-commerce.
Keep in mind that when evaluating backplane speed, the packet balancing occurs only as fast as the uplink. Expandability occurs only through a cascading network of very expensive, large switching chassis—and an additional layer of devices is required to achieve full redundancy. Furthermore, most switching solutions do not contain viable wide area network, high-availability load-balancing features for extended networks.
Routers and Caching Systems
iTCM products complement routers and caching systems. An iTCM product, for example, can offer additional expandability, availability, and security above basic routing and caching functionality. The product can also reverse-manage the load balancing between cache servers and routers, further improving performance. This was the case in the network design implemented by NASA a few years ago, as demand for information on the John Glenn "return to space" mission grew at a feverish pace. NASA employed a dynamic configuration—combining an appliance-based iTCM product with routers and cache servers—to meet user demand.
Turnkey load-balancing appliances are "network appliance" software/hardware products that offer full IP support and easily enhance traffic performance. These appliances are typically deployed in redundant pairs and placed between the server farm and the network—operating jointly as parallel and hot-spare iTCM devices, as shown in Figure 2 . This redundancy offers fail-safe, cost-effective operation and significantly minimizes maintenance. Servers can be upgraded and managed without any downtime or effect on the network.
Figure 2. Internet Traffic and Content Management Deployment
One key advantage of an appliance-based iTCM solution is that it provides OS independence, enabling the organization to implement any type of application or Web server into the mix. Also, the design approach of an iTCM appliance offers a stable balance between high functionality, speed, dependability, flexibility, and cost-effectiveness. Since an iTCM appliance is a software solution combined with a hardware platform, it can be upgraded easily.
For institutions that need continuous e-commerce and secure connections as well as intelligent application interaction, these products are valuable. Also, iTCM appliance solutions are primarily subsets of, and not replacements for, applications whose functionality is highly distributed across large clustering systems.
One example of an appliance-based iTCM solution is PowerApp.BIG-IP, a new appliance that Dell Computer will make available in October 2000. PowerApp.BIG-IP utilizes software—BIG-IP—licensed by F5 NetworksTM , Inc., and a 2U server hardware platform developed by Dell. The BIG-IP software offers a load-balancing solution that provides high availability and intelligent load balancing of Web traffic and content.
Choosing the Right iTCM Solution
Several criteria for selecting the right iTCM product for your business follow.
The ability of an iTCM product to eliminate a single point of failure determines its dependability. A common way to achieve dependability is by offering failover capability through a redundant pair of iTCM devices. PowerApp.BIG-IP, for example, uses a method called session state mirroring to enable failover in less than one second—with uninterrupted service to customers. For instance, if a customer is transferring a large file via File Transfer Protocol and PowerApp.BIG-IP fails over from the active to the standby controller, the file transfer will continue—uninterrupted. Session states are mirrored in RAM, so no end-user sessions are lost.
Optimal Quality of Service and High Availability
To best meet your networking demands, you should look at the scope and sophistication of load-balancing intelligence within each iTCM product. Load-balancing options should include numerous traffic distribution algorithms—such as round robin, round trip time, and packet rate—and QoS features that track and gain intelligence based on current conditions, thus improving performance with each request.
Specifically, an iTCM product should detect errors and reroute traffic automatically by actively querying content accuracy and application performance. By performing application-layer (layer 7) testing, the connected user is assured that the iTCM product has thoroughly checked for all the different processes involved in creating a dynamic page before routing service requests to the server. This testing eliminates the occurrence of error messages because of overloaded servers, software failures, or bad or missing content.
iTCM products should also provide intelligent persistence features, such as SSL session ID tracking, to ensure that users stay connected to a single server while completing a transaction. This feature is invaluable for handling transactions from user environments such as AOL, where numerous users can be assigned the same IP address. This multiple assignment of IP addresses can confuse an iTCM product and concentrate traffic on a single server in the server farm. Instead, your iTCM product should read specific session IDs from an SSL transaction, ensuring the user is uniquely identified and delivered accurate and timely content until the transaction is complete.
Another key requirement to ensure high availability is to offer functionality known as traffic prioritization. This function enables the network to offer varied access service levels based on traffic source, type, or destination—guaranteeing access. For example, rules can be established that always give priority to credit card transactions or to content from or to a specific domain. Traffic prioritization provides the most flexibility and further optimizes availability to a broad range of business applications.
Similarly, the product should be able to drill down and identify specific types of traffic based on HTTP header information. This capability gives an e-business greater control over a wider range of traffic, because the business can see more granular levels of traffic details at the application layer. For maximum flexibility and control of traffic, the product should be able to recognize and provide intelligent, high-availability load balancing to any HTTP header, including HTTP version, HTTP host field (also known as URL), and the HTTP method used by the request.
The product also should offer SSL acceleration, which is especially critical in e-commerce applications. SSL acceleration is a function used to off-load SSL processing from servers, enhancing their performance while improving response time for and traffic management of customer transactions. SSL acceleration improves the performance of e-commerce servers and provides security, speed, and traffic management during business-critical online transactions—from a single location—without the cost of installing additional hardware or software on each server.
Since the iTCM product is typically placed in front of the server farm, it must provide enhanced security measures to protect against common attacks and to route traffic around hacked servers. It should also mask the well-known ports of the actual servers being load balanced to prevent unauthorized access to these ports.
Vendor Credibility and Support
When choosing a vendor for a load-balancing product, knowing the vendor's customers and the vendor's history of customer service are important factors. The vendor should provide proven examples that showcase e-commerce, high customer-traffic loads at varying times of the day, and distributed networks across multiple continents.
Technical support is another critical evaluation area. Vendors should have a strong reputation and investment in tech support, and offer on-site installation and training of their system. Seek customer references and look for specialized technical expertise in load balancing or iTCM, as opposed to broad networking support. Your diligence in understanding these areas will ensure that your installed system will be optimized for your application and traffic requirements.
It Is Your E-Business —So Choose Carefully
Continually enhancing your Web application's availability, reliability, and speed are critical to optimal Internet QoS. Today's users who want to purchase products or services on the Internet will quickly choose a competitor if they are faced with less than optimal performance, or missing or bad content on your Web site. These shortcomings can quickly result in lost business, lost brand equity, and lost market share. To help you choose the right load-balancing product for your Web infrastructure, consider the following questions:
- How does your company ensure 24x7 availability of hardware, software, and applications?
- What metrics are you using to quantify the cost of downtime or erroneous responses?
- How are you testing applications and links prior to deploying them?
- How is your company measuring, and then forecasting, for network growth?
- How do you handle system overload problems?
- How do you accommodate the need for additional servers or other network capacity?
- Describe your company's growth rate.
- What effect does taking a server off-line have on your clients?
- What amount of downtime do you experience when you add capacity?
- What are the single points of failure in your network?
Carefully assessing these questions as they relate to your business can simplify your task of selecting a load-balancing product to manage your traffic needs for today and tomorrow.
David Barclay (firstname.lastname@example.org) is a product marketing manager for Internet Server Products in Dell's Enterprise Systems group. Prior to joining Dell, David was a product marketing manager at Compaq® Computer Corporation in its Desktop and Portable PC divisions. David has a B.B.A. and an M.S. degree in Marketing from Texas A&M University.