How HPE & Arista Are Leading the Data Center of the Future

HPE and Arista bring you a shared vision around deliverying secure Hybrid IT solutions built on industry-leading software-defined infrastructure.

Provisioning network and security services often takes weeks or even months, thanks to legacy data center networks characterized by complex, hardware-defined and static physical (CLI based) configurations and multiple silos of management.

Image 1.png

Now, modern data centers are being transformed with more efficient architectures and technologies, including compute, storage, and networking coming together as Hybrid IT. That’s why your organization requires network architectures that are open and programmable—and that are completely aligned and integrated into data center technology stacks, including compute, storage and cloud.

In partnership with Arista Networks, HPE offers a software-defined, cloud-first network architecture designed to help you address transformation challenges.

HPE and Arista share a common vision for data center networking around the need to deliver secure Hybrid IT solutions and experiences built on industry-leading software-defined infrastructure to help you operate workloads with speed and agility to grow your business.

The HPE and Arista partnership provides you with best of breed networking solutions that are superior to legacy networking solutions and that are complementary to HPE compute, storage, virtualization and cloud offerings.

Arista provides a networking foundation in the data center

The HPE Stack for Hybrid IT is open at every layer, is highly programmable and automated and allows customer choice to use best of breed to components. Arista provides a best-in-class software-defined networking layer that enables automated provisioning and programmability through virtualized network service delivery.

You can choose to modernize and scale these elements independently or consume them as converged offerings Either way, you can take advantage of new compute innovations, storage data services and software-defined networking fabrics are a critical part of a Hybrid IT foundation.

Arista’s best-of-breed technology was recently validated by its “Leaders” position in Gartner’s 2017 Magic Quadrant for Data Center Networking, validating the continued market acceptance of Arista EOS®, cloud management, telemetry, containerized EOS (cEOS)™ and how Arista is leading the way you can build cloud-scale networks.

image 2.png

https://www.arista.com/en/company/news/press-release/3500-pr-20170710

Arista products are being integrated into our HPE data center infrastructure. Our joint focus is on addressing our customers most pressing data center and cloud networking use cases, including highly virtualized data centers, IP storage, high-performance computing, and data center security, to name a few.

image 3.png

With HPE and Arista, we bring you the world’s best-selling computer partnered with the industry’s faster growing data center networking vendor. The HPE and Arista portfolios brings architectural advantages and unprecedented scalability around our common vision: to deliver secure Hybrid IT built on industry-leading software-defined infrastructure to help you operate workloads with speed and agility to grow your business.

Learn more

HPE and Arista Network Solutions

HPE Trusted Network Transformation

HPE and Arista Software-defined, cloud-first network architecture

Related:

Fast Forward with the Right Networking Solution for Your IP Storage

If you’re looking for scalable, agile and automated solutions for your IP storage deployments, make the move to HPE and Arista—and take advantage of easy to deploy and manage networking solutions with lower TCO.

HPE Networking_Arista_fast forward_blog.jpgIt is universally accepted that you buy data center infrastructure to run your business applications. It is natural that you expect that infrastructure to ensure continuous delivery and high performance of applications.

With that in mind, let’s take a look at networking solutions from HPE and Arista that can help ensure the success of IP storage deployments in enterprise data centers.

Once upon a time, when Ethernet’s top speed was one gigabit per second, debates raged in the networked storage community about the viability of Ethernet as a transport of storage traffic. These debates faded away as 10GbE Ethernet followed by 40GbE and 25/100GbE became a reality driven by the Ethernet economics of scale. Today, enterprises are deploying IP storage in their data centers and, as I mentioned, you expect networking infrastructure to ensure high availability and high performance of business applications.

From HPE and Arista: Reliable, agile networks designed with IP storage in mind

Your concerns stem from the fact that storage networks are fragile and require special attention. TCP/IP is a best effort guaranteed delivery protocol that will get congested due to many reasons, including retransmissions, storage devices timing out and Ethernet speeds mismatches. The lack of adequate visibility of networking flows does not help. HPE and Arista have networking solutions that address your concerns and deliver reliable high-performance networking that gracefully handles bursty storage traffic and networking speeds mismatch. HPE and Arista deliver just that giving customers

Successful IP storage deployments require networking infrastructure that delivers reliability with no packet loss, deterministic behavior with robust error recovery and predictable performance that handles bursty storage traffic and networking speed mismatches.

HPE and Arista offer spine/leaf switch solutions that are designed for IP Storage and similar applications. The Arista 7500R spine switch, the Arista 7280R family of leaf/spine switches and the Arista 7020R leaf switches offer low latency, deep buffers, lossless Ethernet and a rich set of analytics tools for to monitor and remedy networking congestion.

  • Low-latency lossless networking—Switches port-to-port latencies oftentimes get more attention than warranted. When it comes to applications performance, aggregate latencies across the entire data path matter more than the latency of one device. Switches are certainly not the slowest element in the data path, but it is great to have low latencies. HPE and Arista switches offer some of the lowest latencies on the market. When coupled with lossless Ethernet based on Data Center Bridging (DCB), low latencies make a big difference and give customers an extra edge over competition that may be deploying slower and more traditional solutions.
  • Deep buffers with VOQ architecture—The truth is that most, if not all, data centers contain server connections with disparate speeds ranging from gigabit all the way to 25GbE. On the uplinks side, speeds go up to 100GbE creating speed mismatches that need deep buffers to smooth data traffic especially in IP storage environments with large files. Arista 7500R, 7280R and 7020R all come with extraordinary large deep buffers that enable them to significantly reduce congestion. Virtual Output Queue (VOQ) available on all above switches helps in eliminating head-of-the line (HOL) blocking by maintaining separate virtual queues for each output port. Deep buffers with VOQ enhance networking and applications performance.
  • Networking monitoring and analytics—Network congestion is a reality of life. Monitoring and analytics tools have tried to detect and alleviate congestions, but with less than desirable success. Arista has taken an intuitive approach to detecting congestion by taking snapshots of network flows in microsecond intervals. The change to smaller intervals allows Arista’s Latency Analyzer (LANZ) to detect microbursts and TCP incast conditions in real time. Arista’s Rapid Automated Indication of Link-Loss (RAIL) feature uses the switch intelligence to proxy send TCP Session Resets (RST) to facilitate recovery of nodes that have failed or lost connectivity to networking switches. Arista’s LANZ and RAIL help reduce the impact of congestion on networking performance and reduce failover times to enhance network and data center availability and reliability.

A closer look: HPE and Arista solutions for IP storage

HPE and Arista offer you end-to-end data center solutions that include servers, storage and networking. The HPE offering is supported by HPE Pointnext expert service teams with local presence worldwide. HPE gives you a single worldwide contact for all data center infrastructure, support and services. The HPE infrastructure offerings include HPE blade and rack servers, HPE 3PAR StoreServ flash storage, Nimble storage and Arista networking solutions.

If you’re looking for scalable, agile and automated solutions for your IP storage deployments, look no further than HPE and Arista—and take advantage of easy to deploy and manage networking solutions with lower TCO.

Arista_IP Storage Sales Play_Blog Slides_May 20 2017_FINAL.jpg

Learn more

What’s the future of Networking?
Get inspired at Enterprise.nxt.
> Go now

Related:

  • No Related Posts

HPE Distributed Cloud Networking: Policy-Based Automation for Cloud Networks

In this blog, let’s look at how HPE Distributed Cloud Networking (DCN) is able to enable cloud-ready data center networks.

Distributed Cloud Network .jpegOur customers have often approached me seeking a solution to a traditional network which takes days or weeks to change to respond to the business. I hear your concerns about these kinds of issue:

  • I need to implement a network policy to block port 23.
  • I also need to redirect traffic from a specific IP subnet to a firewall.
  • I need to provide network connectivity between my Virtual Machines, Docker Containers and Bare Metal Servers.
  • I need to implement micro-segmentation for Virtual Machines and Docker Containers on the same subnet.
  • Can the network team set a QoS traffic class for applications that use port number 8080?
  • I need these network policies to follow the Virtual Machine if it migrates to a different host.

The solution? HPE Distributed Cloud Networking (DCN)

DCN is an overlay network virtualization solution that does not require you to rip and replace your existing network infrastructure. HPE DCN allows you to implement your network policies in minutes—instead of days or weeks. In addition, the HPE DCN solution also includes HPE Virtualized Security Services (VSS) functionality to:

• Prevent malicious traffic on the network through segmentation and policy enforcement
• Detect malicious traffic on the network through visibility and security monitoring
• Respond to malicious traffic by implementing dynamic security automation and locking down the source of malicious trafficData centers NOT cloud ready.jpg

Still in doubt? Interested to see HPE DCN in action?

Check out HPE DCN demos posted on the HPE YouTube channel:
HPE Data Center Networking playlist

How about a customer case study which uses HPE DCN?
Liberating the cloud storage world: Key Information Systems builds an object
storage offering with HPE cloud, server and networking solutions

Related:

  • No Related Posts

IMC’s Flexible Choices – Multi-Vendor Network Management

Network Management - IMC.jpgYou need the right IT tools to resolve any issues you’re faced with. When administrating, managing and monitoring wired and wireless networks, you need a tool that delivers full Fault, Configuration, Accounting, Performance and Security (FCAPS) functionality.

HPE offers two enterprise-caliber Software Network Management System options, IMC Standard and IMC Enterprise, plus a comprehensive range of modules that expand the functionality depending on your specific management requirements

IMC Standard

IMC Standard provides features and functions that are designed for comprehensive management of HPE and 3rd party network infrastructure, and can be expanded to support tens of thousands of nodes.

IMC Enterprise

IMC Enterprise provides the same functionality as IMC Standard but includes the optional NTA (Network Traffic Analyzer) module and WSM (Wirelesss Services Manager), it also can be expanded to support tens of thousands of nodes and with its hierarchical architecture, you can deploy a “Manager of Manager” configuration (the management and network visibility of subordinate IMC servers and the devices across the subordinate IMC server instances)

IMC Optional Modules

Network Traffic Analyzer (NTA)

One of IMC’s most popular optional modules (this module is included with the IMC “Enterprise” platform), NTA provides visibility into the applications that are running on your network and who is using them. NTA acts and a sFlow/Netflow/NetSteam collector and helps unlock the value of these native device features.

Application Performance Manager (APM)

Application Performance Manager (APM) module enables IMC to monitor the servers and applications running on them. , allowing administrators to visualize and measure the health of critical business applications and their impact on network performance. Please see the link below for more information about the APM module.

Service Health Manager (SHM)

The Service Health Manager module allows users to define, monitor and report on Service Level Agreements (SLAs). The SLAs can be defined based on anything IMC is monitoring including from NTA and APM. SHM also allows the configuration and monitoring of NQA and IP SLA (protocol based pings) to monitor real application performance on the network including Jitter, latency etc.

Business Service Performance (BSP)

BSP allows users to model business critical services and displays their performance on a state of the art dash board. It can also incorporate the performance of both physical and virtual customer Data Centers. BSP leverages the IMC Platform, APM, SHM and NTA to provide a comprehensive insight into the performance of business critical services.. Please see the following link for more information about the optional IMC BSP module.

VAN Fabric Manager (VFM)

The VFM module simplifies the management of data center and Fiber Channel (FC) storage area network (SAN) fabrics. HPE also added analytical functionality to this module called HPE FlexFabric Network Analytics which, in conjunction with the HPE FlexFabric 5950 Switch, delivers real time visibility of microburst network congestion which negatively impacts overall network operations and performance. Please see the following link for more information about IMC’s VFM module.

Branch Intelligent Management Software(BIMS)

BIMS provides zero touch deployment and ongoing management of CPE (Customer Premise Equipment) and other network infrastructures via the TR-069 protocol. Please see the following link for more information about the optional IMC BIMS module.

Quality of Service Manager (QoSM)

IMC’s QoSM module enhances visibility and control over QoS configurations across the network and helps administrators focus on service planning and optimization of networking resources by providing a rich set of device and configuration management functions.

Intelligent Analysis Reporter (IAR)

IMC ships with a large number of pre-defined reports. However if the administrator wants to define their own reports they can use IAR. IAR allows administrators to define and import their own report templates into IMC, giving them access to the full range of rich information contained in the IMC DB, and in the format and structure they want.

IPsec VPN Manager (IVM)

For VPN management the IMC VPN Manager (IVM) delivers a total VPN solution, which allows you to construct an IPSec VPN network and provides the ability to effectively monitor the operation and performance of it with a graphical VPN topology, in real time, to troubleshoot VPN issues quickly.

MPLS VPN Manager (MVM)

For customers looking for Multiprotocol Label Switching (MPLS) VPN management, this optional IMC module features include VPN auto-discovery, topology, monitoring, auditing, resource allocation, and performance evaluation.

Remote Site Manager (RSM)

RSM is a light weight version of IMC that can be deployed on remote sites to keep management traffic local and solves issues around managing through NAT firewalls. RSM provides status and other information to a central IMC Standard for centralized remote site management.

Software Defined Networking Manager (SDNM)

IMC’s SDN module, provides management of HPE SDN Controllers and visibility into the Openflow configuration across the network.

Service Operations Manager (SOM)

The HPE IMC Service Operation Manager (SOM) module that focuses on operations and management flow to provide full IT lifecycle management adhering to ITIL v3.0 including services like policy design, operation, and configuration improvement.

TACAS+ Authentication Manager (TAM)

This TACACS+ (TAM) module provides authentication, authorization, and accounting functions for network devices or IT users in order to deliver network device management security.

User Behavioral Analysis (UBA)

The UBA module provides network administrators the ability to audit network usage and behavior of internal users and can be used to help solve security challenges.

VAN Connection Manager (VCM)

The VCM module helps automate the deployment of virtualized applications, removing the network as the change bottleneck. VCM delivers a template, profile-based approach, for managing network configuration policies. Administrators define policies in a template which is then applied to the ToR Swtich port automatically when a VM is deployed or moved.

Wireless Services Module (WSM)

For new Aruba Wireless customers, it is recommended that you consider investing in Aruba’s AirWave management platform for Wireless managment. However, for mixed Aruba and Unified Wirelss WSM provides comprehensive wireless management.

Resource Automation Manager (RAM)

The Resource Automation Manager Module provides network fabric orchestration tools for service application delivery. This optimizes network resource utilization for specific cloud-based or virtualized applications or tenants while accelerating the deployment of applications.

Unified Connection Health Manager (UCHM)

The UCHM module monitors the health of networks deployed with Microsoft Lync. The software allows for monitoring of Lync-associated hardware like servers, PTSN Gateways, and Lync client endpoints in real-time.

Interested in learning more about HPE Networking Management Systems, Solutions & Services? CLICK HERE

Related:

  • No Related Posts

HPE and Arista Networks at Discover 2017

Announcing HPE’s next-generation compute experience at HPE Discover 2017 in Las Vegas

On June 5, HPE will be addressing our customers’ demand for a ‘no compromise’ Hybrid IT experience! We will announce new software-defined infrastructure innovations that deliver unparalleled agility, security and economic control from the core to the cloud to the intelligent edge.

Whether you will be in Las Vegas or watching live around the globe, our HPE DCIG will be hosting over 37 demos, 40 sessions, and will be highlighted on mainstage during the Day 1 General Session.

For you “networking” types wanting to learn more about how HPE and Arista networking solutions fit into this news – I would encourage you to attend or view these networking specific demos and sessions below!

HPE and Arista sessions_demos.png

https://www.hpe.com/events/discover/

Live from Las Vegas.png

Related:

  • No Related Posts

What is Real Time Network Analytics and Why Customers care about it

In today’s world of hyper speed business decisions and need to be agile enough to stay ahead of the competition, match or exceed new market demands, and manage demanding customer expectations, your customer facing application behavior will succeed or fail based primarily on its performance.

This puts a tremendous responsibility on Network Engineers (aka Network Operators) in Enterprises and HyperScale Data Centers as most of the unexplained application performance shortfalls result from an underlying network infrastructure not providing proper performance and scale on demand, or it was not adequately designed for optimum application performance. The Network Operators will have to ensure that their networks are responsive, “always on” and capable of meeting the ever-growing demand from applications they run. Providing operators with deeper instrumentation and telemetry data about the network help operators diagnose network issues, plan and fine-tune the network to provide improved performance and make optimal use of network resources.

One of the main causes of these unexplained application performance issues stems from latency caused by underlying network congestion. Among the causes of a network, congestion is an elusive type of network congestion called “Microburst.” As the name suggest “microbursts” are sub-second periods of time when major bursts of network usage occur at line rate and can temporarily overflow the switch buffers and cause packet loss or backpressure.

Traditionally “congestion” has been associated with switch ports being utilized at close to line rate. In a congestion scenario, packets can be dropped by the switch or flows may backpressure due to lack of buffer space. However, a more recent analysis has uncovered the existence of these “microbursts” occurring more frequently than we may have guessed and there was no good way to detect them, resulting in network engineers looking for the proverbial “needle in a haystack” to find the causes of unexpected application performance issues in their network.

Typically, these “microbursts” do not last long enough to be detected by traditional switch counters such as SNMP or port statistics. This is because traditional tools used to monitor network traffic patterns, such as RMON and SNMP, have been based on a polling model where data is typically collected at one second or longer intervals. What about the events that will occur within these polling intervals? With the evolution to 100GbE attachment in the data center, within even a one-second interval a 100GbE interface could go from idle to forwarding over 280 million packets and back again. In a traditional SNMP/RMON polling model this 280 million packet burst can become invisible.

Let’s look at the potential business impact of these microbursts in a High-Frequency Trading (HFT) environment:

  • In a NetworkingWorld article, Charles Thompson – NI manager of system engineering, stated “When trading floors open at 9:30 am Eastern time, their networks are flooded with a ridiculous number of trades that have been queued up since the night before. To analyze performance issues, network managers often have to break out a one-second period into smaller, microscopic intervals. So, they’ll chop up the one-second interval into 100-millisecond intervals, 10-millisecond intervals, or 5-microsecond intervals for investigations. When you get to a sub-second resolution, it’s referred to as a microburst. It’s a small period of time when a major burst of usage occurred…….But, we’ve had many customers requiring 100-microsecond increments, who will take advantage of this drill down capability.”
  • An InformationWeek article stated, “A 1-millisecond advantage in trading applications can be worth $100 million a year to a major brokerage firm.”

Networks are critical to the business as they deliver applications and services to the rest of the organization. Networks must have high performance, low latency, reliability and security. Network/data center downtime is expensive and impacts the business outcome. By proactively detecting these elusive networks microburst allows the network operators to run their network at the most optimum performance level.

c05347793.png

Learn more about the HPE FlexFabric 5950 100G TOR (Top of the Rack) Switch. This switch will provide the capability of detecting these elusive microbursts by embedding BroadviewTM Instrumentation analytics in the switch.

Related:

  • No Related Posts

Accelerate Data Center Transformation with new HPE Arista Consulting, Support and Products

Learn how HPE expands our Arista Networks partnership to help customers transform to Hybrid IT environments with consulting and data center networking solutions.

On March 6, Hewlett Packard Enterprise continued to expand our strategic Arista Networks partnership with the introduction of a host of comprehensive, new HPE Technology Consulting, Support Services, plus the addition of innovative new Arista HPE Networking Data Center Transformation_blog.jpgswitching solutions. These new offerings are targeted at helping you transition to hybrid, software-defined and cloud data centers.

Ready for Hybrid IT transformation?

Getting started on this journey can be quite daunting and many organizations would benefit from having a trusted advisor to help them navigate though the complexity. HPE and Arista are uniquely positioned to help your organization transition to Hybrid IT enviroments.

In order to reduce transformation risk, HPE offers a very mature methodology which can take you on a guided approch with a focue on necesarry steps to make this change. We account for technology, people, process and ongoing operations. We help manage risk while helping you transform your operations while keeping your business running.

Regardless of your starting point, HPE can help you develop and execute a transformation through a series of pragmatic steps, applying our proven HPE Trusted Network Transformation methodology to mitigate the risks of change.

HPE’s full portfolio of consulting services around all Arista products includes:

  • Lifecycle Services: Strategy | Assess | Design | Integrate | Optimize
  • Global capabilities to integrate the Arista portfolio into customers environment
  • Experts to demystify, advise and articulate the value of Arista’s Software Driven Cloud Networking

TSC.png

HPE is also introducing a portfolio of Arista Network Support Services that go beyond Arista’s one-year limited warranty and service offering. Our portoflio includes: Startup Services, Foundation Care services, Proactive Care services and Datacenter Care.TS.pngHPE IT Services deliver unmatched depth and breadth of expertise with global reach—all while leveraging your existing investments. Whether you’re an industry leader or a growing startup, we’ll help accelerate your IT transformation journey to drive business further.

Powering mission critical, software-defined data center workloads

In addition to these new support offerings, HPE has been aggressively integrating Arista networking into HPE Data Center Infrastructure Group (DCIG) solution offerings. We’ve been collaborating on addressing many of your most pressing data center and cloud networking use cases, including highly virtualized data centers, IP storage, high performance computing (HPC) and data center security. Let take a closer look:

Virtualization is unprecedented demands on the data center, accelerating the evolution of new application architectures and is having profound effects on the network.

  • HPE and Arista offer a comprehensive portfolio of virtualization solutions and services that span HPE ProLiant ML/DL300, DL500 or Apollo 2000 System servers, VMware vSphere and Arista 7050X and 7250QX 10/40GbE wire speed switches—purpose built to address the growth of web, cloud and dense virtualized multi-tenancy environments.

IP Storage: The cost of deploying/maintaining traditional storage networks is growing exponentially. New requirements for compliance and new applications such as analytics drives ever-increasing volumes of unstructured data collection.

  • HPE and Arista bring together a best of breed SDS architectures to support emerging software-defined cloud network and IP Storage based use cases. These HPE solutions span across our HPE StoreVirtual, ProLiant, Blade, Apollo Servers, Scality Ring and Arista 7280E 100GbE deep buffers switches optimized for and lossless spline/leaf storage networks.

High Performance Computing: HPC is continually increasing in size, density, power requirements, storage, and performance. As these systems grow, the network becomes more and more important to allow full utilization of compute resources. Many HPC environments tend to neglect managing the network.

  • HPE and Arista are collaborating on HPC architectural best practices that take advantage of software defined networks to scale large cluster growth and low costs and drive high compute efficiencies. Our HPE HPC product portfolio spans our Apollo servers and SGI HPC system along with Arista 7500E in the core and 7150 and 7050X Series switches at the edge to support 10/40/100GbE networks.

Data center security: This encompasses greater network scale, virtualization, big data and analytic security requirements, protecting critical data and end-user privacy and assuring business continuity. This in turn is leading to a demand for a new approach in network and data security.

  • HPE continues to offer innovative new Arista products and solutions such as Arista AlgoMatch™, an advanced network enforcement solution for modern cloud networking along with the new Arista 7160 Series25/100GbE programmable switch which provides a flexible and scalable solution for access control, telemetry and secure cloud networking.
  • Arista Data ANalyZer DANZ 2017support for the Arista R-Series Universal Leaf and Spine platforms, bringing improved visibility to 25G and 100G networks. DANZ, powered by Arista EOS® combined with AristaCloudVision® for automation and telemetry, delivers the hyperscale visibility platform required to secure today’s cloud centric applications and workflows.

To learn more about the HPE Arista Networks partnership, check out the inforamtion below. Or contact your local HPE sales team or channel partner.

HPE Arista resources:

Data sheets:

Related:

  • No Related Posts

HPE and Arista accelerate customers transition to a software define-data center

Last month at the HPE Global Partner Conference, Antonio Neri announced that Arista Networks would become HPE’s preferred networking partner for customers embracing software-defined infrastructure and cloud. This new strategic partnership expands and complements our current data center networking portfolio while also helping accelerate our customer’s transition to Hybrid IT and next-generation software defined cloud data centers.

Starting today– HPE customers and partners can purchase Arista Networks data center switching products directly from HPE. Learn more about these solutions by visitinghttps://www.hpe.com/us/en/networking/data-center.html

Picture9.jpg

Why is this good for HPE customers?

HPE and Arista share a common vision around the need to deliver secure Hybrid IT solutions built on industry-leading software-defined infrastructure helping customers to operate their workloads with speed and agility to grow their business. This partnership will provide our customers with best of breed networking solutions that are superior to legacy networking solutions and that are complementary to our HPE Data Center Infrastructure Group (DCIG) solutions including HPE compute, storage, virtualization and cloud offerings.

Supercharge Your Data Center with HPE and Arista Networks –SL11887

If your headed to HPE Discover in London – you can join HPE’s Dom Wilde and Paddy Power Betfair’s Richard Haigh to learn more about this exciting new partnership and how HPE and Arista can help you design a simplified, higher scaling and more automated software-defined data center.

Not attending Discover London?Tune in to the live streamstarting November 29, 2016 to watch all of the general sessions with Meg, spotlight sessions, innovation sessions and digital executive interviews.

What other data center networking solutions are we showcasing at Discover London 2016?

HPE 25/100GbE Networking Solutions – demo #11447 –ourHPE serverand networking teams have joined forces with solutions that spans HPE servers, network adapters, transceivers, cables and Ethernet switches designed and tested to deliver a comprehensive 25/100GbE networking ecosystem. View our25/100GbE Infographicto see if these solutions are a fit for you!

HPE Network Virtualization – demo #11086 and #11087 –solutions that help make networks as readily consumable as compute resources. HPE Distributed Cloud Networking(DCN) and VMware NSX help datacenter operators manage a distributed, multi datacenter environment in a simple, open and agile way leveraging SDN and network virtualization.

HPE Ethernet VPN– demo #11084 –EVPN has emerged to offer a strong end-to-end solution for datacenter VXLAN networks. This demo will showcase howHPE EVPNenables multi-tenancy and application mobility in your data center.

Arista Networks Software-defined Cloud Networking – demo #11213 –The HPE-Arista partnership advances customer capabilities in maximizing the value of their networks in supporting mission-critical workloads with simplified management and provisioning, rapid service delivery and lowering total cost of ownership.

xxx.png

https://www.hpe.com/events/discover/

Related:

  • No Related Posts

Why are Real Time Network Analytics so important?

By: Gautam Chanda, GPLM DC Networking, HPE

In today’s world of hyper speed business decisions and need to be agile enough to stay ahead of the competition, match or exceed new market demands, and manage demanding customer expectations, your customer facing application behavior will succeed or fail based primarily on its performance.

This puts a tremendous responsibility on Network Engineers (aka Network Operators) in Enterprises and HyperScale Data Centers as most of the unexplained application performance shortfalls result from an underlying network infrastructure not providing proper performance and scale on demand, or it was not adequately designed for optimum application performance. The Network Operators will have to ensure that their networks are responsive, “always on” and capable of meeting the ever-growing demand from applications they run. Providing operators with deeper instrumentation and telemetry data about the network help operators diagnose network issues, plan and fine-tune the network to provide improved performance and make optimal use of network resources.

One of the main causes of these unexplained application performance issues stems from latency caused by underlying network congestion. Among the causes of a network, congestion is an elusive type of network congestion called “Microburst.” As the name suggest “microbursts” are sub-second periods of time when major bursts of network usage occur at line rate and can temporarily overflow the switch buffers and cause packet loss or backpressure.

Traditionally “congestion” has been associated with switch ports being utilized at close to line rate. In a congestion scenario, packets can be dropped by the switch or flows may backpressure due to lack of buffer space. However, a more recent analysis has uncovered the existence of these “microbursts” occurring more frequently than we may have guessed and there was no good way to detect them, resulting in network engineers looking for the proverbial “needle in a haystack” to find the causes of unexpected application performance issues in their network.

Typically, these “microbursts” do not last long enough to be detected by traditional switch counters such as SNMP or port statistics. This is because traditional tools used to monitor network traffic patterns, such as RMON and SNMP, have been based on a polling model where data is typically collected at one second or longer intervals. What about the events that will occur within these polling intervals? With the evolution to 100GbE attachment in the data center, within even a one-second interval a 100GbE interface could go from idle to forwarding over 280 million packets and back again. In a traditional SNMP/RMON polling model this 280 million packet burst can become invisible.

Let’s look at the potential business impact of these microbursts in a High-Frequency Trading (HFT) environment:

  • In a NetworkingWorld article, Charles Thompson – NI manager of system engineering, stated “When trading floors open at 9:30 am Eastern time, their networks are flooded with a ridiculous number of trades that have been queued up since the night before. To analyze performance issues, network managers often have to break out a one-second period into smaller, microscopic intervals. So, they’ll chop up the one-second interval into 100-millisecond intervals, 10-millisecond intervals, or 5-microsecond intervals for investigations. When you get to a sub-second resolution, it’s referred to as a microburst. It’s a small period of time when a major burst of usage occurred…….But, we’ve had many customers requiring 100-microsecond increments, who will take advantage of this drill down capability.”
  • An InformationWeek article stated, “A 1-millisecond advantage in trading applications can be worth $100 million a year to a major brokerage firm.”

Networks are critical to the business as they deliver applications and services to the rest of the organization. Networks must have high performance, low latency, reliability and security. Network/data center downtime is expensive and impacts the business outcome. By proactively detecting these elusive networks microburst allows the network operators to run their network at the most optimum performance level.

Screen Shot 2016-09-28 at 10.06.38 AM.png

Learn more about the HPE FlexFabric 5950 100G TOR (Top of the Rack) Switch. This switch will provide the capability of detecting these elusive microbursts by embedding BroadviewTM Instrumentation analytics in the switch.

Related:

  • No Related Posts

Performance management in IMC – Part 2: Cool stuff

In part 1 of this series we saw what we can do in terms of performance monitoring without any customization in IMC. This means only by installing IMC and discovering devices … Now let’s see the customization part.

Here is where the cool stuff begins:

  1. We can add some performance views with the data we already have, in order to enable. For example, here are some Trend Lines of CPU and Memory usage for our core devices:
    Perf_Advanced_picture_1.png
  2. We can also add some customised “Real-Time monitoring” pages (in Performance Management à Real-Time monitoring). Doing so we can have a very quick access to live information for live debugging.
    Perf_Advanced_picture_2.png
  3. You may feel frustrated with only a few performance metrics to monitor … Don’t worry, there are about 445 predefined performance indexes you just need to activate in order to monitor these 🙂
    Perf_Advanced_picture_3.pngSo we can, for example, monitor device temperature, interfaces bandwidth, error packets, QoS … Below is an example of adding temperature monitoring:
    Perf_Advanced_picture_4.pngAnd the result:
    Perf_Advanced_picture_5.png
  4. And if you need even more … You can add your own global index (everything you want if this can be obtained through SNMP), either in an existing category or in a user-defined one. Only things you need to know are the SNMP system Object Identifier of the data you need and the formula to calculate the performance index. Below we add SFP transceiver Transmitted Power for ArubaOS devices to the predefined category “Transceiver Current Transmitted Power”.
    Perf_Advanced_picture_6.pngAnd so we can now add this metric as we usually do with predefined ones 🙂
  5. Let’s finish with another interesting feature. We can add any of the monitoring indexes configured above to the topology. For example, we can say that for switches we want to display the temperature and that we want to display some interface statistics on the links:
    Perf_Advanced_picture_7.pngAnd so when you select a device or a link you should see something like this:
    Perf_Advanced_picture_8.png

There are many other things you can do, such as adding dashboards, playing with RMON, configuring schedule reports to be sent by email … I’ll let you discover by yourself. And also we can use IMC NTA (Network Traffic Analyser) module in order to analyse more precisely the traffic on our infrastructure. I promise I’ll be back soon with another blog in this series.

netmanfabe

Related:

  • No Related Posts