WAN Optimization

Transparent Accelerators Transparent accelerators use similar metrics as nontransparent accelerators to identify peer devices on the other end of the network (automatic discovery or static configuration) but do not manipulate packet headers for data in flight. This helps to ensure compatibility ... [full story]

Nontransparent Accelerators Nontransparent accelerators are designed to optimize traffic that has been redirected to it and explicitly forward the optimized traffic to a statically defined or dynamically discovered peer through an explicit connection to that peer accelerator. In this way, the ... [full story]

Architecture of Accelerator Services All accelerator devices (and acceleration technology in general) have some posture to services deployed in the network. These network services include any action that can be performed against network packets and flows, including network visibility, monitoring, end-to-end ... [full story]

Integrating Accelerators into the Network To understand how accelerators coexist with the foundational network technologies for visibility, control, and resource alignment, it is also important to understand how accelerators integrate into the network. Integrating accelerators includes deploying accelerators and delivering to ... [full story]

TCP Optimization TCP is particularly challenged in WAN environments due to the connection-oriented, guaranteed-delivery behavior of the protocol. Furthermore, TCP generally has only a limited amount of memory capacity assigned to each connection, meaning only a small amount of data can ... [full story]

WAN Optimization WAN optimization is a set of services that overcomes the performance limitations caused by transport protocols, network conditions, and network utilization. The three most common WAN optimization services employed by accelerators include TCP optimization Data suppression Compression Many accelerators offer WAN optimization capabilities ... [full story]

Overview of Accelerator Technology Network accelerators provide a combination of functions to help improve the performance of applications over a network. These functions help to mitigate many of the performance-limiting factors in enterprise networks today such as those discussed in previous ... [full story]

Understanding Accelerator Control Features and Integration The previous sections in this chapter outlined a few of the many key network technologies that you can employ to help understand how the network is used and to effectively align network resources with the ... [full story]

Post-Queuing Optimization Traffic that has been released from a queue by the scheduler according to queuing policy and shaping then passes through a series of post-queuing optimizations. Post-queuing optimizations help to ensure optimized delivery of packets across the network by interleaving ... [full story]

Queuing and Scheduling Once traffic has been classified (identified) and pre-queuing operators have been applied (policing, dropping), the router then queues traffic for service onto the next-hop link. Queuing is defined as the way a node temporarily stores data while waiting ... [full story]

Differentiated Services DiffServ is more commonly used than IntServ and is referred to as soft QoS due to its reliance on per-hop behaviors at each node in the network, dictated largely by a common understanding and configuration of how to handle ... [full story]

Integrated Services IntServ is commonly referred to as hard QoS due to its ability to set flags related to reliability, bandwidth, and latency and also its reliance on the Resource Reservation Protocol (RSVP) to signal and reserve the desired QoS for ... [full story]

Packet Marking Packet marking involves the manipulation of a 1-byte (8-bit) field within the IP packet header called the type of service, or ToS, byte. Network devices use this byte of data to determine the relative priority and handling requirements of ... [full story]

Pre-Queuing Operators Once packets have been classified, the next set of functions in the QoS behavior model, pre-queuing operations, are employed. Pre-queuing operations are employed before queue capacity in the network device is consumed. Pre-queuing operations ensure that queue capacity is ... [full story]

Packet Classification Packet classification allows a network device (such as a router, a switch, or even an accelerator) to partition network traffic into multiple priority levels or classes of service based on a variety of match conditions. These match conditions help ... [full story]

Employing Quality of Service The previous section presented two means of examining the network to gather a fundamental understanding of how the network is being used. This information forms the foundation from which you can make decisions about how to align ... [full story]

Viewing Network Utilization Many network administrators today still do not know what type of traffic is consuming network capacity, what the top applications are, and who the top talkers are. Most routers, switches, and other network devices today include feature sets ... [full story]

Network Interface Cards There are two common methods to improve the performance of the NIC. These methods involve offloading of the network traffic from the processor of the server, and allowing hosts to place data directly into other hosts' memory. The ... [full story]

File System Considerations A file system facilitates the storage needs of an application, housing any data that the application or server itself cannot hold in physical RAM. A file system is a creation of its parent, the operating system. File systems ... [full story]

Fiber Channel Fiber Channel (FC) attached storage is commonly connected to the server via Lucent connectors (LCs) embedded within a small form-factor pluggable (SFP) optical adapter, such as a PCI host bus adapter (HBA). HBAs have throughput ratings of up to ... [full story]

Serial Attached SCSI (SAS) Serial attached SCSI (SAS) is the next generation of SCSI interconnect where communication with the drives occurs in serial, not parallel, between the controller and attached drives. With serial operation, clock rates can be increased due to ... [full story]

Disk Storage The spinning disk is inevitably going to be the choke point for application performance in many server environments, as information is exchanged between the server bus and magnetized areas on a spinning platter. Typical storage deployments have changed radically ... [full story]

Random Access Memory Memory selection for servers is a constantly evolving process. Although naming is very similar between types, the performance and capabilities differ significantly. Common server memory implemented today includes double-data-rate (DDR), synchronous dynamic random-access memory (SDRAM), double-data-rate two (DDR2), ... [full story]

Support for 32-Bit and 64-Bit Architectures Some processors support a 32-bit architecture, whereas others support a 64-bit architecture. As operating systems evolve toward a greater array of 64-bit support, applications must also support the ability to communicate with a 64-bit processor. ... [full story]

Hyper-Threading Hyper-threading and Super-threading are offerings that are specific to processors manufactured by Intel Corporation. Hyper-threading allows a single processor to appear as if two separate processors exist. Super-threading allows a thread, or thread of execution, to be shared with a ... [full story]

Dual-Core Support A dual-core processor is simply a single processor chip that contains two processors that can work in parallel. A common benefit of a dual-core processor is that the individual processors will each have their own primary cache, or Level ... [full story]

Front Side Bus Speed The front side bus is a significant factor in the server's ability to move data about the server. The numerical rating of a server's front side bus designates at what speed the processor can transfer data to ... [full story]

Cache Capacity Layer 2 and Layer 3 caches reside in the processor chip. These caches accelerate the processor's ability to receive data. Processors today support caches that range in size from 1 MB to 24 MB or larger. Typically, the larger ... [full story]

Central Processing Unit There are several factors to take into consideration when selecting a processor. As the CPU speed war continues, competition will only benefit your applications and their performance. The continual evolution of processors will make selecting the correct processor ... [full story]

IBM IBM (http://www.ibm.com) offers a broad array of 32-bit and 64-bit servers. These servers support operating systems offered by vendors such as Microsoft and Red Hat. Select servers include the POWER5 series processor that is designed to specifically support operating systems such ... [full story]

Hewlett-Packard Hewlett-Packard (http://www.hp.com) offers several different server platform product lines and a proprietary operating system called HP-UX. In addition to supporting HP-UX, the Hewlett-Packard server platforms support other operating systems, such as Microsoft Windows Server 2003, Novell Corporation's NetWare, Red Hat ... [full story]

Red Hat Red Hat (http://www.redhat.com) offers two versions of the Enterprise Linux operating system for data center computing, Linux AS and Linux ES. Red Hat is an operating system vendor that supports an array of processor platforms, including Intel Corporation's Itanium ... [full story]