NimBUS for Web Monitoring
Summary:
The NimBUS for Web Monitoring solution monitors web page (URL) load
times and page contents. The solution monitors end-to-end response
times for web-based applications and business services. Additionally,
the solution monitors web server processes, resources and web server
network connectivity response times. All web status data is analyzed
and displayed in graphical alarm dashboards, performance reports
and SLA reports.
Download
our Web Site Monitoring Solution Overview Whitepaper
Internal and external end-users can only be as
productive as the performance of the web-based applications and
services they interface with. Poor web-based application availability
and performance can also dishearten first-time and repeat customers
- both conditions, if allowed to persist - can impact the business
bottom line and place the business at risk.
The NimBUS solution provides comprehensive web
server and web application monitoring to ensure a high quality end-user
and customer experience.
Primary Features:
Additional Features:
Web page load response time monitor and
web page content monitor
The web browser is becoming the sole client of
choice in today's computing paradigm. Nearly all application data
now "passes through" a web page and is accessed via a
URL. Given that the access to a URL is dynamic and does not depend
on a pre-established connection, a user will not know if a particular
URL is unavailable until he or she has attempted to access it. If
it does exist, is it downloadable in a reasonable time? Finally,
is the content correct?
From a user experience point of view it is vital
that the content is available, quick to download and correct.
The NimBUS URL Response probe can be set up to
monitor many URLs for the 3 key characteristics:
- Is it available?
- Does it download in a reasonable time?
- Is the content correct?
The probe monitors availability by checking http
status codes e.g. error 404. For download times, the expected time
to download can be set up as a general value for all pages and specifically
overridden for particular pages if required. Checking for content
can be used in any way that is suitable but is generally used to
ensure that no deliberate or inadvertent corruptions of the page
have been made. As an example, checking on correct header and footer
text can be used as a generic content check for all pages.
Web page load response
time monitor
The NimBUS URL response probe will test access
to defined URLs and record the amount of time it takes to fully
load each page. The response probe is extremely light weight and
can be strategically deployed to monitor URL load times from multiple
points of perspective. The probe allows for immediate testing during
configuration - page load response times will be displayed during
the on-demand test and can then be used to establish a baseline
threshold for alert definition and generation.
Figure below: The NimBUS URL response probe will
collect and archive response time poll values, and automate performance
report generation.
Web page content monitor
The NimBUS for Web Monitoring solution utilizes
the URL response probe (discussed above) to also monitor web page
content. The probe is easily configured to watch for text strings
contained (or not contained) within defined URLs. The probe allows
for immediate testing during configuration, this will ensure the
probe is properly detecting the defined text string before deploying
into production. Alerts can be generated when the defined text string
is absent or when a defined text string is present. Note that the
probe supports monitoring web pages with proxies and user authentication.
Figure below: The NimBUS URL Response Probe can
be easily configured to monitor URL page contents. Poll values will
be analyzed for alert generation and can be archived for performance
and SLA trend reporting.
Web application response
time monitoring from an end-user's perspective
The NimBUS for Web Monitoring solution can measure
and monitor performance of web-applications from the users' perspective.
The solution can be used to gain insight into overall latencies
incurred by users who are trying to get real work done and who have
no concern for the inner workings of the web application or its
delivery components.
Web response monitoring for business-unique,
simple to complex applications
The NimBUS solution supports response time monitoring
for business-unique, end-to-end web-based application configurations.
The solution is uniquely designed and positioned to quickly adapt
and monitor practically any simple to complex web-application environment.
For simple to complex web-based applications
and services, the NimBUS solution can record, playback and monitor
the response time for each of its distinct phases. For example,
a web-based application consisting of four distinct phases may be
as such; 1) web-application launch, 2) web-application login, 3)
web-application database query, 4) web-application logout - in this
example each phase is a unique end-user facing experience, which
may likely encompass a unique underlying technology infrastructure,
and to effect optimal end-to-end web application performance, each
phase should be monitored as such.
This complete and partitioned approach towards
web application response time monitoring allows for determining
the total end-to-end web application response time as well as isolating
particular phases of the web application that may have less than
desirable response times.
Figure below: NimBUS end-to-end response time
(rt) measurements may be taken large-grain or fine-grain; overall
monitoring from start to finish, or monitoring and measurement of
each phase of the overall interaction with the end-user. Phases
might include: Login, data query, data load, and logoff.
For each web-application phase, as well as the
total end-user web-application experience, response time data is
analyzed in realtime and displayed in graphical alarm consoles.
Additionally, response time data is archived for performance trend
reporting and SLA-based response time monitoring and reporting.
Web server monitoring
to ensure integrity and peak performance
The NimBUS for Web Monitoring solution includes
a specialized web server resident probe; the probe will be platform
specific. The specialized probe targets a wide array of web server
metrics to ensure status awareness for Web Server Administrators.
Examples of web server resources and system processes monitored
include: CPU, memory, disk space, event log files, web specific
processes and services, and much more.
The solution's fully integrated performance monitoring
and reporting functionality provide the necessary server visibility
for performance degradation analysis, trend detection to preempt
problems, and enables proactive server resource and capacity planning.
Web server connection
monitor
The NimBUS for Web Monitoring solution will test
the amount of time it takes to connect to a designated web server
on a network - the solution will issue scheduled network connectivity
tests against defined web server IP addresses or web server domain
names, i.e. www.conversesoftware.com. The result of this network
connectivity test will reveal web server availability and web server
round trip connectivity response time.
Figure below: An example NimBUS performance report
showing web server connectivity response times during a one day
polling period. Archived performance data can be viewed by day,
week, month, or quarter.
Web application
and service-port connection monitor
The NimBUS solution will test the availability
and connectivity response time for critical web services, i.e. HTTP,
FTP, and others. The solution will issue scheduled connectivity
tests against defined web-server service ports with alerting and
reporting on availability and round trip response time results.
Figure below: NimBUS Alarm Manager with an example
dashboard showing a cluster of web servers and critical web service
ports being monitored.
Alarm Dashboards
The NimBUS for Web Monitoring solution displays
all alerts in the NimBUS Alarm Manager dashboards. Alarm dashboards
are extremely flexible to allow for developing views that suit personal
and business preferences.
Figure below: Here we have an example NimBUS Alarm
Dashboard displaying URL response times from various geographic
locations. In this scenario the response time data from the collective
locations is consolidated and analyzed against a national response
time SLA (see 'SLA-based web response time reporting' below for
more details on NimBUS SLA reporting).
Performance monitoring
and reporting
QoS (Quality of Service) data resulting from NimBUS'
periodic polling can be archived for future performance reporting
and trend analysis. Archived performance data can be viewed by day,
week, month or quarter.
Figure below: The performance report below is
one of many out-of-the-box web specific performance reports that
are available. In this example web-based performance metrics are
being plotted over time. Here we are showing URL page load response
time in seconds, and for the duration of one month. All web status
poll values that have been discussed in this NimBUS for Web Monitoring
overview can be archived and displayed for performance trend analysis.
SLA monitoring and reporting
NimBUS offers an advanced SLA-based performance
monitoring and reporting solution whereby all alarm and performance
data collected and processed by NimBUS are also analyzed against
NimBUS Service Level Agreements. The NimBUS Service Level Manager
product provides for SLA creation, monitoring, and reporting. With
the NimBUS SLM solution it is possible to have web specific SLAs
defined, monitored and generating compliance/breach reports in a
matter of minutes to hours.
Figures below: The SLA screen sequence below is
one of any number of NimBUS SLA monitoring and reporting examples.
With NimBUS SLM we have defined a SLA for nationwide
web page response time - URL response time tests will be performed
from each geographic location. Each geographic location represents
individual service level objectives that are defined within the
collective SLA (SLO definition: URL load time for each location
should be 3.3 seconds or less). When service level objectives are
breached the SLA will risk falling out of compliance. Early warning
alerts will be generated as the SLA begins to degrade.
In the example below URL response time monitoring
is being performed from San Francisco, Anchorage, Seattle and New
York. Each city represents a SLO and the objective for each SLO
is to have URL load times (at each polling interval) that are less
than or equal to 3.3 seconds - the goal for each SLO is 99% compliance.
San Francisco has breached its SLO enough times to fall below compliance
- as of now, and at the current rate, the collective breaches for
San Francisco have not, and will not compromised the national SLA.
If the breach rate were to accelerate the national SLA could be
compromised.
Historical reports shows SLA compliance during
each monitoring period and allows for investigating conditions for
each previous period. Clicking on the 'History' tab will show SLA
compliance and breach history.
SLOs within the SLA can be selected to view its
performance data against defined thresholds. Selecting San Francisco
in the top level SLA will launch a performance report and highlight
URL poll intervals that have exceeded the defined threshold of 3.3
seconds.
The NimBUS SLM solution caps off our comprehensive
web monitoring capabilities - deploying the full compliment of realtime
alerting, performance data archiving and reporting, and SLA-based
performance monitoring ensures that web environments will be scrutinized
in proactive and pre-emptive fashion, and most importantly, in the
context of business requirements and end-user expectations.
The NimBUS for Web Monitoring solution requires
minimal effort for implementation, training, and will not require
extensive ongoing maintenance.
If you would like to request a free trial of NimBUS,
please click here. For free white
papers or additional information click
here.
|
