Technology Communities |
NetworkingMaterial on this page is intended solely for the purpose of content review by SNIA members. Tutorial material may be read and commented upon by any SNIA member, but may not be saved, printed, or otherwise copied, nor may it be shared with non-members of the SNIA. Tutorial managers are responsible for responding to all comments made during the open review period. No responses will be given to comments made outside the open review period. Jump straight to an abstract:
The Abstracts
Fibre Channel Technologies: Current and Future The objectives of this tutorial are:1) Provide the user with a Primer on Fibre Channel a) Flexible, Scalable relative to Topologies, Speed, Performance, Distance, Node connectivity and Low costb) Communication and Data Overhead (Framing, Data Communication, Latency, Efficiency, Routing Control, and Access Control),c) Redundancy, Availability, and Failover, d) Applicability in SAN with large IT User Base2) Project the market outlook and roadmap of Fibre Channel 3) Share what is New in Fibre Channel Standards for Protocols APIs, and Management. a) Management And Ease Of Use b) Operational Flexibility and Scalability and Security.
Learning Objectives
SAS & SATA Combine to Change the Storage Market Serial Attached SCSI (SAS) allows systems to be built that accommodate large numbers of either SAS and/or SATA hard disk drives.This presentation, intended for OEM, System Builders and End-Users, describes the capabilities of the SAS interface, overview of it’s technical capabilities, how it’s designed to interoperate with SATA drives and when combined, how these technologies can be combined to deliver some very compelling storage solutions. Also included will be an update of the new capabilities included in the SAS 2.0 specification.
SANs Across MANs and WANs Extending storage networks across distance is essential to BC/DR (Business Continuance/Disaster Recovery), compliance, and data center consolidation. This tutorial will provide both an overview of available techniques and technologies for extending storage networks into the Metro and Wide area networks and a discussion of the applications and scenarios where distance is important. Transport technologies and techniques discussed will include SONET, CWDM, DWDM, Metro Ethernet, TCP/IP, FC credit expansion, data compression, and FCP protocol optimizations (Fast Write, etc). Scenarios discussed will include disk mirroring (both synchronous and asynchronous), remote backup, and remote block access. Learning Objectives
Fibre channel Over Ethernet (FCoE) FCoE is a concept that encapsulates Fibre Channel frames into Ethernet Frames and amalgamates these technologies into a network fabric that can support Fibre Channel protocols and other Ethernet based protocols (such as TCP/IP, UDP/IP etc.). The presentation will show the general FCoE concept and describe how it might be exploited in a Data Center environment and its position with regards to FC and iSCSI.The requirements on the Ethernet Fabric for support of FC protocols will also be shown.The benefits of converged I/O (storage messaging, clustering, etc.) to the Data Center will be discussed. The state of the FCoE protocol (Standards position, probable adoption projections, etc.) will also be presented. Learning Objectives
Ethernet Enhancements for Storage: Deploying FCoES A key challenge that the network and storage industries are now facing is the abundance of new high speed interconnect protocols proposed for future data center applications. What customers really want is to reduce the number of disparate networks and minimize the complexity surrounding the management of their data centers. In this presentation we take a peek into what the future may hold for high speed fabrics and investigate the potential for their unification. We will provide a market and technical overview of the competitive landscape for next generation 10GbE technologies with particular focus on the operational characteristics and implementation aspects of Ethernet. The audience will learn how hardware, software, standards and innovations driving the vision of a unified fabric utilizing 10GE for the data center of the future. Fibre Channel over Ethernet (FCoE) is a key technology that allows IT managers to transport Fibre Channel (Storage) traffic over a standard 10GbE network. There are tremendous benefits for implementing FCoE in terms of reducing cost and power of deployed networks. The session will provide an end-to-end view of implementing FCoE and evaluate its benefits from a host and switch perspective. Learning Objectives
Comparing Server I/O Consolidation Solutions: iSCSI, InfiniBand and FCoE This tutorial gives an introduction to server I/O consolidation, having one network interface technology, to support IP and storage (Fibrechannel, SCSI) applications. The benefits for the end users are discussed: less cabling, power and cooling. For these three solutions, iSCSI, Infiniband and FCoE, we compare features like Infrastructure / Cabling, Protocol Stack, Performance, Operating System drivers and support, Management Tools, Security and best design practices. We also highlight the implications for the Datacenter front end network, bridging / routing in the access / aggregation / core layer, exponential increase in number of MAC and IP addresses. Learning Objectives
NAS and iSCSI Technology Overview Requirements for additional storage are booming. There is a strong trend that most of today’s storage will be consolidated and 'networked'. This presentation shows the different approaches to 'Storage Networking’ and discusses file and block I/O related subjects. Topics discussed will include:
Learning Objective
SMB2 - Big Improvements in the Remote Filesystems Protocol SMB2 (“Server Message Block” 2) is the most widely-used remote file system protocol.As SMB2 has evolved from CIFS (Common Internet File System) and SMB, the protocol has expanded from Microsoft Windows client and server operating systems, to implementation by many other server OS versions with file and print capability.This includes every NAS and Server vendor – some develop their own implementations, and some adapt the widely-used open-source Samba program.(There are also CIFS/SMB clients for non-Windows operating systems.)Much of this development work was done without benefit of adequate or current protocol documentation, thus requiring tedious work involving observation, packet inspection, error injection, and laborious testing.The release of Windows Vista and Windows Server 2008 introduced a substantially renewed protocol, SMB2.Documentation for SMB2 is publically available (downloadable) on the Web, which has made implementation a much easier task for non-Microsoft operating systems.Interoperability is a priority, and a CIFS/SMB/SMB2 Plugfest at the SNIA Storage Developers Conference in Sept—08 provided significant testing and interaction to demonstrate this. What does this mean to system designers and End-Users?SMB2 was designed to support a much wider range of network transport speeds and latencies than CIFS or SMB.Network speeds ranging from kilobits/second to gigabits/second, and latencies ranging from <1 millisecond to hundreds of milliseconds are all supported.SMB2 is also much less “chatty”, possibly reducing the need for “accelerator” devices.These protocol features not only allow clients and servers to communicate more effectively (and with less overhead) over wired LAN and WAN connections, but also support increasingly ubiquitous Wireless connections, which are sometimes less reliable, especially when clients move around. The presentation will close with performance data over a variety of link speeds and latencies to provide insight into how well SMB2 performs when compared to SMB version 1. Learning Objectives
Scaling NFS Through pNFS I/O requirements of NFS clients in Grid Computing and HPC-Technical Computing are rapidly growing beyond what NFS servers can supply them. This problem can be solvedby increasing the NFS server's bandwidth. This tutorial explains the pNFS modifications to the NFSv4 protocol to scale NFS servers. This is enabled by increasing the aggregate bandwidth possible to a single file system through multiple NFS server endpoints serving data in parallel to a single or a cluster of NFS clients.We also explore how hetrogenous storage servers and layout protocols, e.g., files/blocks/objects can be made transparent to NFS clients through a common open systems architectural framework. We explore how clustered file systems can be fully parallelized through pNFS without the need of host based agents or stubs. We conclude with few prototype examples and use cases along with the current status of the protocol development. Learning Objectives
PCI Express and Storage
System IO Architectures are now PCI Express. As multi-root IO Virtualization is being defined, shared IO infrastructures are on the horizon. This session describes PCI Express, Single Root and Multi Root IO Virtualization and discusses the impacts of all these changes on storage connectivity, storage transfer rates, as well as the implications to Storage Industry and Data Center Infrastructures. Learning Objectives
Fabric Consolidation with Infiniband In the era of exploding datacenter requirements for networking and storage services, and with the increasing power, space, and budget concerns over the infrastructure, fabric consolidation becomes inevitable. InfiniBand was designed from day one for fabric consolidation. With 120Gb/s links and with ultra low-latency characteristics, InfiniBand provides a well provisioned foundation for consolidation of networking and storage. Additional features such as QoS, partitioning, virtual lanes, lossless fabric, and congestion management facilitate true consolidation of fabrics along with connectivity of InfiniBand islands to Ethernet and Fibre Channel clouds through gateways. This session highlights the features for fabric consolidation and the various protocols that run over InfiniBand with emphasis on storage protocols. Learning Objectives
Six Sigma Performance Analysis for SAN Switches
Traditionally SAN switch performance is measured in average and maximum throughput.Do these metrics capture an accurate data profile of SAN switch performance?I believe there is a better way to measure performance and predict bandwidth needs for capacity planning.Six Sigma has long been used for measuring and predicting production in manufacturing plants.In SAN we are in the business of producing data throughput with a finite amount of bandwidth.Six Sigma can then be applied to SAN traffic to understand and predict SAN performance patterns with 99.73% confidence, which leaves less bandwidth under utilized and increases your SAN ROI. Learning Objectives
|
Login
Featured Events
|
