Verizon deploys Highlight reporting in Asia

Verizon Business has significantly extended its contract with NetEvidence by rolling out Highlight performance reporting to customers in eight countries throughout the Asia Pacific (ASPAC) region. Highlight's Software-as-Service (SaaS) capability lets Verizon's Dedicated Internet Access customers get real-time, browser-based performance reports on their Internet Connection - with no hardware or software to install. The information includes Traffic Levels, Trends, Availability, and (on compatible sites) analysis of the Applications running over the connection.

"This development shows how easy SaaS makes things for the Network Provider community" said Andi Willmott, NetEvidence's Business Development director. "Normally, deploying capabilities like those in Highlight throughout an entire new region of the world would require significant investment and Hardware and Software changes from the Service Provider, and take a long time. In this case, Verizon simply amended contracts and added Highlight to the relevant pricelists. As the SaaS solution provider, NetEvidence takes care of everything else. Highlight fully supports users in multiple time zones, and Verizon customers with diverse networks across multiple countries can now monitor their performance from a single, intuitive screen."

The change makes Highlight available in Australia, Singapore, Taiwan, Korea, Hong Kong, Japan, India and New Zealand; Interested Verizon customers should contact their Verizon Business account manager.

Juniper Support added to Highlight

The latest version of Highlight exploits the Performance Monitoring capabilities built in to Juniper routers.

Highlight's Performance Insight functionality has to date proved extremely popular with users - for example, Enteprises with Managed Networks who need to keep tabs on end-to-end performance ; or Network Providers whose Service Management teams need to prove to customers that Service Level Agreements (SLAs) are being met.

To date, Performance Insight has only been available on Cisco-powered networks. Using that company's IPSLA functionality, Highlight users can set up performance tests in seconds, monitoring network delay and application performance through Highlight's simple graphical interface, and even use Highlight's News Insight feature to send them email alerts if performance thresholds are breached.

Version 6.6 of Highlight extends this PI functionality to networks built from Juniper hardware, using the vendor's Realtime Performance Manager (RPM) feature. Highlight users running Juniper networks can now configure end-to-end tests between any two points on the network, quickly and simply.

"We're pleased to be extending our support for Juniper" said Richard Thomas, CEO of NetEvidence. "An increasing number of Enterprise customers want it, and our Service Provider partners want to include it in bids. Following our philosophy of keeping things simple, Highlight automatically detects Juniper hardware and correctly configures it for performance testing. The user just clicks the button, and can be receiving Highlight's fast, clear performance reports within minutes."

Additional support for Juniper hardware is in the Highlight Roadmap for 2009. If you have specific queries about Juniper support, call your Service Provider account manager or, if you buy Highlight direct, call NetEvidence's Sales team.

New Face: Mark Bonner

Latest recruit to join the NetEvidence team is Mark Bonner, who takes up the role of Development Manager. Mark was previously managing development at Panasonic Europe and brings a wealth of experience in design methodologies and development practice, as well as many years of experience with both Delphi, .Net and Web Design technologies.

INDUSTRY NEWS

SaaS... going from Good to Great

In the current environment, it's no wonder Software as Service (SaaS) is a hot topic: why pay up front for software you'll then have to support, maintain, and fret about? Better to rent the software, sign in, and pay for what you use. But don't take our word for it: see what IDC said on the subject this month.

In a recent report, IDC found that by the end of this year, 76% of US organisations will be using at least one SaaS-delivered product; and by 2010, nearly 45% of those organisations expect to be spending fully a quarter of their IT budget on SaaS applications. A quarter. If you're an Enterprise software user, that kind of figure is pointing out a sea change in where your applications are likely to be coming from over the next few years.

"...the harsh economic climate will actually accelerate the growth prospects for the Software as a Service (SaaS) model as vendors position offerings as right-sized, zero-CAPEX alternatives to on-premise applications." says the report, led by Robert Mahowald, director, On-Demand and SaaS research at IDC.

It's also going to cause some hiccups, as legacy software providers try to move into the SaaS market, and find it's not as simple as it seems. As an example, Phil Wainewright blogs about Sage, a UK company which launched a Beta SaaS version of its product in early January, only to withdraw it a month later while they addressed security flaws.

SaaS is a philosophy NetEvidence have followed from the outset, and consequently we're well aware of the pitfalls. Our security is good enough that Highlight users include banks, and senior government departments. We know that our Service Provider partners love being able to offer Highlight on to their customers with zero impact on their internal IT or Development people; and that their customers love the sign-in-and-go approach. So, with Marc Benioff driving salesforce.com through the $1Bn revenue level this month, reports like IDC's are bound to validate and grow the SaaS market further - watch this space.

Partner Focus: Claranet

Net Evidence talks to Claranet’s Martin Saunders:-

“…the revelation for the customer is that an SLA can be much simpler : delivering the application is what we’re doing, so that’s where we define service levels. Everything else, like underlying network performance or server downtime, is for the provider to worry about. It makes for a much simpler SLA, and a much simpler relationship with the customer. We just need to show real, end-to-end performance in a simple way that customers can relate to their business.”

Read full article >>

NOTES & TIPS

What is "Jitter"?

JITTER is a term found frequently when discussing Service Levels, and everyone knows that a good network has minimal jitter, as well as low delay. But it would be useful to understand this cagey little term better. What can we teach you about Jitter in the sixty seconds available before you get bored with such Technical issues and click onwards? This is supposed to be a business newsletter, after all.

Let's try three things:

1. Network Jitter is what happens when packets are sent into a network at nice regular intervals, and come out the other end at different and unpredictable intervals. Modest jitter is always going to be present in any network, but significant jitter is only a problem for REAL TIME applications such as Voice or video. Normal applications like Browsing, EMail, SAP, CRM, Mission-critical thin-client tools, Citrix, just don't care. Really.
2. At a higher level, Network Delay is what happens when packets take a long time to get from one end of the network, to the other. High Delay will affect those Normal applications, and Variation in Delay - network performance which is fast one minute and slow the next - will drive users crazy, but this is NOT THE SAME AS JITTER. If ten cars set off from London at five minute intervals to drive to Leeds, then Delay is the average journey time, and Jitter is how much the original five-minute spacing has been altered by the journey when they arrive.
3. Exactly WHAT IS a reasonable level of Jitter, or Delay Variation, in a network is almost impossible to state, and depends on how you're measuring things. There are generally-accepted target levels for normal and for real-time applications, and we'll look at these in more detail next month. But if users are happy, there's no point obsessing over numbers. The important thing is to be able to spot CHANGES: a trend upwards in Delay, or a spike in jitter that corresponds to some user complaints or application timeouts. That's why 'spot' checks on these values don't tell you very much, and why at NetEvidence we focus on showing you patterns, or trends in these measurements. To illustrate some of this, here's an example of a connection with good, and bad, variation in delay:
   
   
   Monday is good: all those spots (each of which represents a Delay and Jitter test) grouped tightly around 30 milliseconds: happy users. Friday is very different: delay spread randomly between 20 and 180 mS. This is going to produce very unhappy users, and is good reason to look for issues in the network such as heavy traffic, or congestion.