Cloudware Operations and Management - Guide to Cloud Computing for Business and Technology Managers - page 384

1. Technical part : The technical part of an SLA (so-called service-level
objectives) specifies measurable characteristics such as performance
goals like response time, latency or availability, and the importance
of the service.

2. Legal part : The legal part defines the legal responsibilities as well as
fee/revenue for using the service (if the performance goals are met)
and penalties (otherwise).

Supporting SLAs requires both the monitoring of resources and service pro-
viding as well as resource management in order to minimize the penalty
cost while avoiding over-provisioning of resources.
There are two types of SLAs from the perspective of application hosting.
These are described in detail here.

1. Infrastructure SLA : The infrastructure provider manages and offers
guarantees on availability of the infrastructure, namely, server
machine, power, and network connectivity. Enterprises manage
themselves, their applications that are deployed on these server
machines. The machines are leased to the users and are isolated
from machines of other users. In such dedicated hosting environ-
ments, a practical example of service-level guarantees offered by
infrastructure providers is shown in Table 18.2.

2. Application SLA : In the application colocation hosting model, the
server capacity is available to the applications based solely on their
resource demands. Hence, the service providers are flexible in allo-
cating and de-allocating computing resources among the colocated
applications. Therefore, the service providers are also responsible
for ensuring to meet their user's application SLOs. For example, an
enterprise can have the following application SLA with a service
provider for one of its application, as shown in Table 18.3.

TABLE 18.2
Key Contractual Elements of an Infrastructural SLA
Hardware availability 99% uptime in a calendar month

Power availability 99.99% of the time in a calendar month

Data center network availability 99.99% of the time in a calendar month

Backbone network availability 99.999% of the time in a calendar month

Service credit for unavailability Refund of service credit prorated on downtime period

Outage notification guarantee Notification of customer within 1 h of complete

downtime

Internet latency guarantee When latency is measured at 5 min intervals to an

upstream provider, the average doesn't exceed 60 ms

Packet loss guarantee Shall not exceed 1% in a calendar month