| Lab tech specs |
|
Currently the entire lab network is run from two main servers that run the
distributed applications and database. One of the servers is a dual Xeon 1.2Ghz
with 4GB RAM and 768GB of RAID-5 storage, the other system being a dual Xeon at
3Ghz with 2GB RAM and 1296 GB of RAID5 storage. Both machines are hooked onto a
1Gbps network connection to the rest of the office network. In addition three
backup units take care of running the daily backups. The entire ensemble of
power has been packed in a small footprint housing that takes up one square
meter.
|
|
| Authorization |
|
Every user of the lab network, staff members, partners, and users, are all
authenticated and authorized prior to being granted access. There is no public
access portal.
|
|
| Access anywhere |
|
The lab network can be accessed from both Microsoft Windows and Linux platforms,
through a web based portal and/or through a few stand alone applications that
interface directly with the distributed applications. The web based portal
limits a few areas to read-only access for security reasons.
|
|
| Scalability |
|
The distributed application that forms the lab network was designed with
scalability in mind. Scalability to facilitate a growing or large scale
organization and scalability in response time and processing speed on the actual
servers. The nature of the distributed application allows for the distribution
of the main back-end components and databases over a cluster of servers while
the front-end access provision can be load balanced and scaled over multiple
servers simultaneously. Similarly, the distributed application allows for
multiple level of redundant operation due to its design.
|
|
| Zero Administration |
|
While technically a buzzword, the lab network requires very little daily
administration and maintenance other than the automated cleanup and backup
services that issue the maintenance control over the network. An organization,
regardless of its scale, would not require additional staff to deploy and
maintain a system like this.
|
|
|
| The lab network |
Our revolutionary lab network system has been the focus of attention in and
outside of mantiCORE Labs and many questions regarding it come up so frequently
that we decided to create this 'of interest' page for anyone who's interested in
knowing more about the actual functionality our system offers us and our
partners.
The lab network consists of a number of distributed applications and services
that can be accessed through various means in order to conduct any and all tasks
relating to project management ranging from maintaining notes and technical
specifications up to full automated build-on-demand. The lab network also allows
us to maintain a close teamwork spirit with members of staff and partners which
are located outside of the main environment.
LAB TEAM
labteam @ manticorelabs.com
|
| Everything at a glance |
|
The biggest strength, apart from customizability, are the capabilities of the
lab system to provide updated and current information at a glance. Most
overviews and tracking systems require you to go through various pages of
different portions of information. The lab, however, allows customized overviews
that will display actual information grouped by their logical position within
research and development projects. The way in which the lab network allows
related information to be displayed in a relational way makes all the difference
between having full insight as opposed to losing all oversight. This relational
model of information also allows information and details to be modified or added
in the exact same way. For example, an engineer could look at their overview
page in the morning and have a current list of tasks, defects, memos, and e-mail
at the fingertips, with all information linked to every other aspect of the
larger knowledge base system. This single desk approach has proven to increase
efficiency by well over 20% simply because of the adaptive format in which the
information is displayed.
|
| Communication |
|
The lab network offers many kinds of communication methods that are all using a
centralized storage. The web based discussion board, the NNTP interface, and the
E-Mail list capabilities are all based around a single storage format. This
allows easy and consistent access through means that are defined by personal
preference. The heart of the discussion sections consists of a traditional NNTP
server.
|
| Defect tracking |
|
Defect and bug tracking capabilities form an important part of the virtual lab
and offers many revolutionary features whereby bug reports and defect
descriptions can be directly linked to the technical information and source code
areas. Defects are sorted and managed based on staff roles and development
sections and provide a high degree of automation in the actual tracking
activities. This enables the team to respond to defects quickly and effectively.
|
| Build on demand |
|
The build-on-demand features of the virtual lab allow for the centralized
compiling and building of projects. This automated system runs on a dynamic
schedule but can be controlled by staff members with build privileges in order
to create specific build versions. In addition to the build control the system
provides detailed build reports and overviews in near real time as well as
distributed compilation for multiple platform targets.
|
| Source tree depot |
|
The virtual lab area, which can be reached through the use of a secured virtual
private network, also maintains the full source code depot and tree for the
projects undertaken by mantiCORE Labs. This enterprise level configuration
management system has been adapted to integrate into various others parts of the
lab network such as defect tracking, architecture notes, etc. which essentially
allows the code to be a reference to more background information while serving
the dual purpose of being a reference point in itself for other sources of
information.
|
| Automated documentation |
|
A major and complex prerequisite of proper technology research and software
development is to maintain active and up to date documentation regarding the
architecture, source code, and project details. This comes in the form of
internal confidential information as well as technical documentation provided to
partners. This time consuming and tedious task has been automated in various
ways to fit the different requirements on a per-project basis. Source code and
API documentation is generated on a daily basis directly from the different
relational sources of information in the virtual lab system, which in addition
also provides a plethora of functionality to track known issues, defects, or
inconsistencies between the different bits of information. Likewise,
architectural documentation is generated by forming a relational combination of
details into a single point of reference that can be transformed into a few
different delivery types such as HTML or PDF documents.
|
| Scheduling |
|
Scheduling and project planning details are maintained through the virtual lab
system for all staff members. While this is not a unique capability it offers
various ways to integrate the scheduling information directly into the research
and development pipelines in order to serve as a point of reference. With task
assignments and their dynamic nature the importance of proper scheduling and the
tedious brain twisting to make priorities fit has been automated to a
comfortable extend that still allows direct modification.
|
| Specifications management |
|
Management of technical and formal specifications relating to one or more shared
projects are directly accessible and searchable from the lab network. A
centralized file storage and management component allows specification documents
to be viewed through the web based portal or downloaded in one of many formats
such as Word documents, PDF, or HTML.
|
|
|
|
|
|
| Why it was developed |
|
One of the most compelling reasons that jump started the development of the lab
network was the need for a large scale automated project management and
maintenance paradigm that would relieve us from the tedious tasks in research
and development. While many tedious tasks are extremely important it was
likewise important to have a framework and infrastructure that would not simply
automate these tasks but would augment them at the same time. Another major role
in the development of the lab network was reserved for centralization of
information. Anyone working on any project under mantiCORE Labs is able to use
the lab network as a single source of technical reference.
With team
structures larger than a dozen people the importance of strong communication and
project overview increases almost exponentially. When elements in such team
structures are decentralized it becomes even more important to have a single
point of reference for all project related material. Loss of insight and
overview is a reason many software projects fail or get into technical trouble
in relation to scheduling, maintenance, and enhancement. When the requirements
for the lab network were met an additional round of development was initiated to
enhance many areas. While these enhancements are an ongoing the lab network has
already surpassed the expectations and original criteria for which it was
needed. These developments also have a direct effect on cost control and
efficiency whereby the entire staff and its managers can do more in less time,
thereby cutting out the traditional layers of middle management which are quite
often present in large scale software research and development projects.
|
|
|