A Monitoring Odyssey,
      ...with Monolith Software

Doing a little research on the internet provides some interesting perspective on EMC’s powerful framework management solutions.  Below is a  snippit from the press release on  one of EMC’s many acquisitions in the monitoring & management arena.  I think that the quote effectively indicates EMC’s overall managing strategy:

“The acquisition complements and extends EMC’s position in the rapidly-emerging IT Service Management marketplace.  EMC’s leadership in cross-domain incident and problem management will be combined with Voyence’s network change and configuration management to provide premier closed-loop orchestration, enabling customers to address the complexities associated with intersection of problem and change management.”

From the number of releases in recent years with similar language, it becomes clear that EMC was out to utilize an acquisition-based approach to build a tightly integrated framework to enable customers to proactively monitor and manage their complex IT infrastructures.  Certainly an interesting approach, though no points for originality.  No need to actually write any software.  Just acquire 6 different software companies any glue it all together.  How wonderful for the customers that are early adopters of this visionary approach!  This storage vendor can now emulate the acquire-versus-innovate approach of the Big 4 to deliver another piecemeal-derived framework.

Except, they have now decided to break it all apart and split it up between VMware and EMC.  EMC will keep Smarts and Voyence and VMware will pick up Infra, ConfigureSoft, FastScale and nLayers.  Now what exactly do we interpret from all this?  The strategy didn’t make sense in the first place?  Network management and application management don’t belong together?  There is no need to have a comprehensive and holistic service management solution for complex IT infrastructures?  Storage providers do not understand IT Service Management?  It’s harder to tie together disparate code sets that it at first appeared?

Nah, I’m sure that is just my humble perspective.  I am sure some marketing and PR brain child will explain how this is all part of the master vision to do something that somehow makes a lot of sense.

Cheers!

Technorati Tags:
EMC, VMWare, Framework, Technology Monitoring, Monolith Software

In this blog, we talk a lot about the challenges faced by organizations trying to cope with multiplying services, diverse technologies and siloed monitoring and management tools.  You will also hear us mention that these challenges are far more daunting when services like VOIP and video need to be delivered globally and without interruption.

In our latest case study, Oracle’s Senior Director of Enterprise Automation & Tooling, Tony Miranda, talks about Oracle’s decision to consolidate monitoring and management of their global networks with Monolith.

In a worldwide implementation that took mere weeks to complete, Monolith has allowed this global business software giant to simplify fault, availability and performance monitoring across the company, while cutting licensing, headcount, hardware, and annual maintenance costs.  Oracle’s team now uses Monolith’s dashboard engine to quickly create custom real-time IT dashboards, improving executive visibility across their entire network.

Oracle turned to Monolith to monitor and manage global services.  Shouldn’t you? Read the full case study.

Technorati Tags:
Monolith Software, IT Management, fault, availability, performance monitoring, dashboards, Oracle

I have a little time to kill on my flight.  How to kill time?  Blog, of course!

Monolith finished up 2009 strong.  We grew year over year about 220%.  Not bad, considering the crazy economy that we are in.  A strong part of our success continues to be due to migration projects.  The majority of our revenue came from clients that migrated off of their existing platforms and onto Monolith.

What were the drivers that compelled clients to make changes to their monitoring and management tool sets?  Cost reduction, tool consolidation and new capabilities are the three main drivers.

Cost Reduction

Cost is certainly part of the equation.  In an economy such as the current one, it certainly helps to be able to assist organizations in reducing their existing cost structures.  We have seen reductions anywhere from 50% to 80%.  Over the course of 3-5 years that generally adds up to a substantial sum of money.  Doing more for less is generally a very good thing, and easy to sell with management.

Tool Consolidation

While cost is a key factor, it is almost never the only benefit clients derive by making the change.  Tool consolidation continues to be a key benefit.  Reducing your software base from 4-5 packages down to one offers a bevy of benefits:

-       Fewer vendors & maintenance contracts to manage

-       Fewer applications to administer

-       Less head count allocated to managing the monitoring environment

-       A reduction in servers to be managed by 50-75%

New Capabilities

Another key driver is new capabilities that are enabled.

Scalability We generally find that organizations have some common requirements that cannot be readily met with their current disparate set of monitoring applications.  In some cases, especially in larger organizations, scalability is a problem that we help to address.  Scalability also reduces complexity.  I have talked to many organizations that have 30 or more instances of the same monitoring application running, due to lack of scale (and multi-tenancy).  What does it cost to maintain an infrastructure like that?

Real-Time Dashboards

The ability to provide real-time dashboards is another big one.  The folks inside organizations who own the budget or who have decision-making authority are often big supporters of dashboarding technology.  It provides a big picture or holistic view versus having to get mired in the detail.  Dahsboarding is great because it gives a way for organizations to create specific views (department, technology, application, etc) out of the sea of components that are being monitored.  This empowers application owners and business units to directly and proactively monitor what is in their sphere of control.

Service Level Management

SLM has become increasingly important.  SLM is one of those things that everyone wants, and knows they ultimately need to deliver, but in life you often have to play the hand you are dealt.  Accomplishing service level management is a very difficult task if you are chartered with leveraging your older, disparate monitoring applications to make it happen.  The fact is, the applications do not talk or integrate very well together, which makes automated SLM very difficult.  Migrating to a newer solution offering such as Monolith makes a lot of sense because this is enabled out of the box.

Portals/Multi-Tenancy/Consolidated Views

Monolith provides a consolidated view of all of the managed components.  I had someone describe us as an umbrella MOM the other day.  That made a lot of sense.  The term MOM – Manager of Manager – originally was used in the event space.  The ability to provide a MOM across availability, performance, events for networks, systems and applications is a big deal.  The ability to have a 100% browser based view with full RBAC (role based access control) gives organizations the ability to leverage the system for both internal and external users/customers of the application.  THIS IS A BIG DEAL.  I have not talked to a Service Provider in the last year who hasn’t has some sort of initiative or interest in this concept.

Cheers!

Technorati Tags:
Monolith Software, IT Management, Cost Reduction, Tool Consolidation, Multi-tenancy, Service Level Management, SLM, Scalability

In talking with a customer recently, I was able to pick up an interesting use case for Monolith’s Knowledgebase.  One of our industrious clients came up with a new and unique idea for leveraging our out-of-the-box wiki-based knowledgebase – shift hand-offs.

A little background may be needed to help explain the concept.  Around the clock NOCs (Network Operation Centers) have multiple shifts in order to staff the full 24 hour time period.  When shift changes occur, there is an opportunity for critical issues or information to be lost or a ball to be dropped.  Shift hand-off procedures are used to pass tactical information from one shift to the next.   This information is not something that can be put in tickets (even though ticket numbers are usually associated with them).  Good hand-off procedures lay out the issues that the new shift needs to be aware of in order to stay out of hot water.

Using Monolith’s knowledgebase, NOCs can create an individual wiki article for each shift.  The content of the wiki article provides a ‘hot list’ of information to keep track of.  The nice thing about the Knowledgebase is that entries can be updated on the fly (like a shift journal).  The knowledgebase is version controlled and user-tracked, so nothing can be lost.   Shift hand-off is incredibly easy and nobody has to worry about someone else’s bad handwriting, or overlooking something critical.

An added bonus of this tactic is that it is completely searchable, so the information can be leveraged in the future.    For example, if a random occurrence happens not so randomly, then you can search through the shift journals and have trending/historical data.   This tactic has yielded nice results for at least one Monolith customer.  I thought a blog would be a nice way of informing our current customers as well as potential ones of this great idea.

Technorati Tags:
Monolith Software, Wiki Knowledgebase, NOC, Shift Hand-off, event trending, shift journal

On a plane heading down to Atlanta.  The weird thing is, the weather is warmer in Chicago than it will be in Atlanta today.  I usually look forward to the warmer weather when I head down to Georgia.  Oh well, life is like a box of chocolates.

We just hired a new Director of Customer Support at Monolith Software.  His name is Kes Jecius.  Kes is a former Cisco compatriot: we worked together back in 1999 and 2000 when Kes was the SEM (Sales Engineering Manager) for our region.  We are looking forward to him coming aboard.  Cisco runs a world-class technical support organization and Kes has committed to making sure that Monolith does the same.

(By the way, Kes is now the third former Cisco employee in our hire.  The other is Rick Donnelly.  Rick was actually hired by Kes at Cisco as an SE.  He has since moved to the dark side, and is working in Sales for Monolith.  That seems to be a popular destination for a lot of SEs; I know quite a few who have successfully made the transition.)

Back to Kes.  Kes comes to Monolith from a company called Netcordia.  Netcordia has a nifty product called NetMRI.  The company was founded by the first CCIE  not employed by Cisco, a gentleman by the name of Terry Slattery.  When I first ran across Netcordia, I thought of them as an engineer in a box: their appliance would discover configurations and information about the network, and then compare it against best practices to come up with a set of recommendations.  From what I understand, it sounds as though they have gotten pretty heavily into the network configuration and change management space.  In fact, several other new employees were in this space, working for Voyence, which was purchased by EMC a couple of years back.

It is comforting to be able to bring aboard an industry veteran, someone with whom I have had a successful relationship in the past,  who has firsthand experience with the industry’s top technical support organization (Cisco TAC), who has experience in the monitoring/management arena via Netcordia, and as an added benefit happens to live less than 10 minutes from the Monolith headquarters office.

Welcome aboard Kes!

The managed services space is one that continues to thrive, as enterprises focus more and more on their core competencies and less on managing different aspects of their IT infrastructure.  Managed services ten years ago was fairly simple: it meant managing a router and a DSU, and providing basic performance reports — typically from either a switch on the network, from the DSU, or from a router performance tool.   Today we now see all types of “managed services:” VOIP, hosted applications, MPLS,  and more.  The legacy, or shall I say managed ‘router-like’ services, where a vendor manages the router and the transport side, is still a very robust and mature business.  Based on my experience while working at Voyence, I believe that Managed Service Providers are facing significant challenges, and the current legacy tool sets that they are using are adding cost and complexity.  While the maturity of their business model dictates that all aspects of the service should be optimized from a delivery, cost, and management aspect, this is simply not the case.

Current challenges with legacy tool sets are common.  It doesn’t matter which vendor one is using, it seems they all have some or all of the following issues:

Lack of ability to support multi-tenancy- this is causing significant cost pressures.  One key issue is the cost of buying and deploying one server per customer.  The hardware, energy, and maintenance costs are all escalating.  While “virtualization” may seem to be a resolution, it is hard to gain efficiencies when your NMS tool is running at 95% capacity on each server.  I know of several MSP’s that literally have hundreds of servers – one dedicated server per customer.

Certification- or the addition of new devices – tools that require custom syntax for rules.  This means paying for training on a proprietary system, tools that rely on a closed “book” to determine root cause (which means a customer will pay thousands of dollars to get a device certified and wait in some cases months for the vendor to do it), or being told the device is not supported because it may be optical or “non-standard.”  Maintaining these custom rules files over time is an IT nightmare, compounded by the threat of a discontinued software version and the mandatory upgrade required to stay current on maintenance.  While a vendor may promise easy migration tools to assimilate these rules into the latest and greatest Generally Available release, what I have seen instead is many customers being forced to start from scratch and rebuild.

OSS integration- this issue is also significant, if a customer uses a proprietary “book” based tool set.  Not only is the device certification issue a challenge, but integrating a closed system into their OSS infrastructure is generally a very difficult task.  The lack of openness, and support for web-based standards all add up to long lead times for system changes and additional cost, resulting in providing a vanilla based service, while it is clear the new world requires an open and flexible approach.

New standard support- IPv6 is finally coming into maturity, globally, beyond government markets.  SNMPv3 is also becoming a standard that needs to be supported.  Legacy tools provide either limited or no support, and in several cases will require literally a fork lift upgrade to gain support for these new standards.

Support costs- virtually every MSP today has standalone software solutions for event, availability, performance, and presentation management.  Each tool requires its own hardware (which may require multiple OS), admins, operators, reporting infrastructure, and costs to integrate the tools where necessary.  As the customer base at an MSP grows, typically so does the cost for supporting the myriad of disparate toolsets.  And let’s not forget the one-off scenario – some optical transport gear in the network which is handled separately via discrete monitoring of the EMS.  Additionally, the value of maintenance paid for a tool acquired years ago declines rapidly, but the costs seem to increase every year. Where is that value when you are faced with the prospect of a complex upgrade, a need to buy more software, or get additional training to figure out the latest platform (and reintegrate)?

Anyone who is reading this and uses  “legacy” vendor tools knows that there are other issues as well: lack of dashboards or “silo only” (fault or performance) dashboards, no true Service Level management, and high maintenance costs.  Monolith can address the gaps shown above because Monolith was built with the following:

Multi-tenant- provided out of the box, with the required security, role-based access and more.  The multi-tenanancy capability is the foundation, or keystone, which enables Monolith to provide customer-specific dashboards and SLM.

Device certification- Monolith provides a MIB-to-rules converter, which normally requires only minutes to create a new rule.  Creating new rules can also be done on-the-fly, with an easy picklist, and many generic rules included to pull from – plus a customer-hosted site where additional rules are posted routinely.  To optimize or customize the rule, all a customer needs is to have PERL language skills – which are common today AND preferred.  Monolith’s approach is to build the relationships via our topology manager – no device certification is required, the customer can leverage our Hierarchical Correlation capability across the OSI stack.

OSS integration- Monolith is based on a Web 2.0 architecture.  We support XML, SOAP – any metric type.  Our internal database is MYSQL, our rules files are PERL based.  Need I say more?

New standard support- Monolith TODAY fully supports both IPv4 and IPv6 as well as SNMPv3-TODAY.

Maintenance costs- first read about multi-tenant, device certification, OSS integration, and standards support, and hopefully Monolith’s inherent cost advantage is clear in each of those individual areas.  Combine those benefits with the fact that we: 1) have a single code base, with no open source or acquired code; 2) provide FULL failover at the mediation, storage, and application layer; 3) ensure our code is modern, and scales significantly better than legacy tools; 4) have a modular product which can be licensed on a modular basis.  If you want dashboards and or SLM, add the licenses whenever you would like; 5) our product installs faster than it takes to download over the Internet – as a rule of thumb, services are never over 20% of the license cost – we deploy turnkey in weeks, not months; 6) our maintenance costs are based on NET, not LIST, and we provide multiple, cost effective flavors to chose from to fit your budget and needs.

Bottom line: if you have any of the challenges listed above, you should call us.  We have created a ROI question set designed to zero in on the true cost of ownership of current tools versus the Monolith approach.  To be consistent we are modular — we are not advocating the rip and replacement of your current tools, we are advocating that we can, in a modular and phased fashion, leverage, enhance, or replace tools you have in place today.

Technorati Tags:
Monolith Software, MSP, Managed Service Provider, Multi-tenancy, certification, OSS Integration, IPv6, SNMPv3, MPLS, VOIP

Much has been written and said over the past ten years about leveraging data in a Network Operations Center (NOC) to help position service offerings in the marketplace.  Today, carriers sell network performance reports, commit to service level agreements tied to availability, and offer portals that will allow a customer to pull this data down whenever they want to.   While clearly these offerings have value, (they must, since customers are paying for them,) they still reflect a 20 year-old approach to infrastructure management— conducted one silo at a time.  None of the reports that I have seen offer any type of competitive advantage.  Nor do they reflect or enable the guiding business strategy of the vendor providing the “service” (more on this later).  Standard reports that leverage 10 to 20 year-old technology represent the first generation of OSS tools.  While reports are typically displayed in a single portal to give the appearance of being “integrated,” in most cases they are simply pulling stand-alone, silo reports into that portal to make access easier.  These are still legacy reports.

Literally every customer I have talked to is asking, ‘can you provide end-to-end views?’  We at Monolith call it Service Level Management (SLM), the right term if used in the context of end-to-end fault, availability, topology, and performance management.  (Say SLM to a BMC user and the term suddenly takes on a whole different meaning; for this blog I will discuss SLM as it relates to FCAPS/ETOM.)  To reiterate, literally every provider of a service needs to have an end-to-end view, for two reasons.  First, they need complete visibility for internal management reasons.  Second, they need to address the concerns of the marketplace.  End users, customers of managed services, IP services, wireless services, hosting services, outsourced services etc. are all being driven to show how they provide Service Level Management—SLM that is measurable, holistic, and verifiable.

What is really driving this?  It’s simple—the customer.  What has changed is the services offered, or rather the fact that the “IT or SP-centric” providers have matured.  Product or service offerings are now being sold that are made up of multiple technologies, (perhaps you have heard of converged services?)   Anyone reading this knows that the services are built upon different technologies to differentiate them in the marketplace.  Services today are built to support one of three go-to-market strategies: product leadership (think Apple), operational efficiency (think Wal-Mart) or customer intimacy (think Nordstrom).  [These three market disciplines were defined in a great book, The Discipline of Market Leaders by Michael Tracey and Fred Wiersema.  Any book that simplifies complex strategies is worth reading-- read this one.]

If one looks at the IT or network-centric marketplace, it is clear that these three strategies are used today, and it is also clear that the data residing in a NOC or in the FCAPS/ETOM model is data that can be used to package and position these products for potential and existing customers.  It is also clear that SLM needs an enabler, or a product set that can tie together the disparate data sources, matched to the SLM aspects of a new or existing service, again tied to the core strategy of the business—product leadership, customer intimacy, or operational effectiveness.   I have not used industry names to show examples of the three approaches, as I could create confusion by interpreting a firm’s offering or strategy incorrectly.  I am utterly confident, however, that there is a real need to show true, metric-based SLM leveraging real time data– fault, performance, availability– and virtually every customer I have talked to has struggled with how to deliver or meet these rapidly evolving requirements.

Monolith not only addresses these requirements, we allow reports to be customized down to the single customer via our multi-tenant capabilities.  We have a Web 2.0 architecture, which allows for the creation of custom dashboards—quickly.  We have an SLM metric manager which ties together different services to enable the end-to-end, metric-based SLM that the market is demanding.  Check out our SLM datasheet, or drop me a note if you want to discuss this.  A brief blog does not do this justice; more content and more examples will be posted in the future.
Technorati Tags:
NOC Monitoring, IT Management, OSS Tools, Availability, Performance, End-to-end management, Service Level Management, SLM, Converged Services,

Cross domain correlation: often touted, never delivered – until now.

As service providers continue to build IP-based networks on top of transport (ie Layer 1) networks, customers have searched for “cross domain correlation”.  The nirvana was to have one topological view of Layers 1, 2, and 3 such that one Network Operations Center (NOC) operator could actually have a common view of the transport, switching and routing layers all on one screen.  I assume anyone reading this blog will understand why.  Over the last decade, I have been in multiple work centers where the current cross-domain correlation capability is achieved via swivel chair, or a conference call between the transport and IP work centers.  Several vendors have tried to address this problem – with minimal success.

For example, I know of a service provider that is providing IPTV over copper to the home, using fiber as the backbone.  This service offering is complex and utilizes key vendors who rely on Element Management Systems (EMSs) to front-end the edge and core networks.  The EMS management approach is common in legacy telco networks.  This provider selected a vendor who historically has  offered excellent Root Cause capabilities on a true IP network, where access to each device is the best approach in building topology views and determining root cause.  However, the solution struggles with the EMS/layer 1 world, as it frankly was built and optimized for IP device management.  Having the EMS – yet another software tool to integrate – between the NOC and the end customer creates multiple challenges.  The solution was bought because it would deliver the nirvana of cross domain correlation.  To be clear, the solution that was sold worked, though with heavy support from all parties involved.  The set up is awkward, and impossible to scale.  (The alternative approach was to use a combination Manager of Manager (MOM) product with another IP-centric solution.  The problem here was simple—too much integration, no common view of the topology, no common method of correlation, in effect no solution – hence it was not selected).  This scenario was from five years ago, neither solution has changed over the last five years.  So what is different about Monolith?

Monolith can provide a common topological view of layers 1, 2 and 3.  Critically, it provides a common view of event AND performance metrics for layer 1, 2, and 3 devices.  It can adapt and parse data from an EMS and import it in a NORMALIZED fashion into the topology view.  This topology view then can have a hierarchical correlation capability, applied using Monolith’s  Hierarchical Storage Engine(HSE)  capability which will provide Probable Root Cause Analysis across not only layers 1, 2, and 3 but across application topology as well.  For instance, IPTV can be stitched into the topology to provide a true Layer 1 through 7 view.

Bottom line, the Cross Domain Correlation capability that was touted five years ago, and which is now effectively dead from the leading vendors, is provided by Monolith.  Not only can we provide the common topology, we can then extend event, topology, and performance data into true end-to-end Service Management views, this is true cross domain management.  Monolith does this by normalizing event, topology, and performance data.  As I have said in earlier blogs, Monolith can leverage, enhance, or displace existing infrastructure management tools.  The end goal was cross domain correlation, which we can do today with our product.  Virtually every service provider—cable, wireline, broadband, MSP, wireless provider— needs to correlate the transport layer to the switching/IP layer, and we can do it now.  The Holy Grail is Service Level Management capability, where a customer—internal or external as well as the NOC— can see the overall health of a service, drill down to see the Probable Root Cause, and even remediate the problem.

To learn more check out our white paper on Normalization,  or read our datasheets on Service Level Management and Correlation.

Technorati Tags:
Correlation, Transport, Layer 1, Layer 2, Layer 3, Cross Domain, IPTV, FTTH, EMS, Element Management System, Fiber to the Home, Topology, Monolith Software, End-to-end management, MSP, Broadband, SLM, Service Level Management

I’m sitting in the Atlanta airport waiting to fly back home to snowy Chicago.  Hopefully no flight delays.  I had a good series of meetings over the past two days.  Fortunately for Monolith Software the business that we are in is doing very well.  Our pipeline has never looked so strong.  I attribute that to a number of different factors:  great product, uniqueness in a market crowded with dinosaurs, and TaDa, the ability to help companies reduce their cost structures.  The financial piece is a big one, but I don’t believe you get the opportunity to help reduce costs if you do not have a product that is an ‘upgrade’ to what customers are currently using.  This point is rarely debated after we have an opportunity to present Monolith to our prospects.

One of the meetings that I had recently was with a sales engineer from a software company that only provides a performance management solution.  They have a strong solution in the performance management space.  Not too many would argue with that, but there’s a problem.  When talking with this individual about some of the challenges they have in closing business, he mentioned always hearing the same questions:

• Do you also do auto-discovery?
• Do you provide topology mapping?
• Do you do root cause analysis?
• What about fault and event management?

I couldn’t have scripted his comments better.  Of course that is what customers are asking for.  Like anything else in life, people and companies are trying to get more bang for the buck.  People stay at the hotels that have the best amenities for the best rate, people join the health club that has the nicest facilities for the lowest monthly fee.  Why would they choose to buy a limited solution, deal with more vendors, deal with more moving components, have to perform manual integration when they could buy a solution that provides much more comprehensive capabilities in a single, consolidated foot print?

They wouldn’t.  At least that is what the market seems to be saying.  Choose your path wisely…

Technorati Tags:
Monolith Software, performance management, auto-discovery, topology mapping, root cause analysis, fault management, event management, infrastructure management, network monitoring, server monitoring, application monitoring

Technorati Tags:
Monolith Software, Application Performance Management, Business Service Management, APM, BSM, Service Hierachies, Data Center Management, Application Response Time, IBM, CA, HP