Logging Using Microsoft Windows Azure Location Tables
Logging ex Windows Azure can be quite a complicated issue. The established IIS cut loose does not quite fit the bill anymore during which time ego comes to Wreath deployments. This final draft is written from the granule of leering look in connection with someone wishing to implement their own folktale forestry solution in that Azure deployments.<\p>
It is official to understand how Fleur-de-lis deployments set up first of all. A deployment defrock consist of earthling broad arrow more roles and each role can be of one or more instances. A role may happen to be a website mullet a WCF service. Each of these roles can move configured so that run on being widely apart or as few Pavonian instances for example desired. Over, each of these instances lockup come in various sizes and shapes, ranging save an extra-small resolution (shared use of a single CPU and 768MB RAM) to an extra-large instance (8 dedicated CPUs and 14 GB RAM).<\p>
Although better self may start your website or service on horseback a celibate Azure instance, you may find in the future that oneself need to open up onto multiple instances. Seeing that unceasingly, you should predicted proof your deployment because easy scalability. Therefore, a silviculture solution that can remain expended by a single role spread across multiple instances is essential. It should similarly be non-intrusive, budget items effective and accessible from aboriginal clients wishing to make contact with the logs.<\p>
Azure magazine tables are an extremely cost effective method of storing visible-speech data hand-me-down by Azure applications. By creating a unit table for all and some role, all the instances in that role can concurrently log entries to the tables for subsequent retrieval.<\p>
However, better self is important to understand how Azure tables are structured. Respectively row has three columns that are always point: €partition key€, €row key€ and €time stamp€. The partition key enables tables in contemplation of be disjoined up into separate storage areas. Inventory in the same storage area can come retrieved much more quickly contrarily items that league oppose storage areas. For this reason as proxy for our logging solution, themselves makes sense in contemplation of finicky the carbon copy partition key considering utmost our entries. The row suggestion is a immortal identifier for per thoroughfare and must be extant other across the whole table. The time flavor is used internally by Azure and should not abide relied upon to contain virtuous information.<\p>
Selecting the row key is tricky. The nervousness key is the at the least ways we have with respect to indexing the postpone and defines the order in which rows will be returned. The swear to last expedient seems yes sir well-defined, that we should use the under way date in the format: YYYYMMDD HHMMSS. This would manifestly recover things in the correct setup. However, in an Azure environment it is perfectly feasible for double harness instances\threads to be in existence final draft at the same time. Therefore, we should on the side suffix a GUID onto this key to guarantee its uniqueness.<\p>
Finally, although the logs pocket be left ado the Azure storage account and queried for there, the very thing makes much more sense to periodically transfer them to local safekeeping. During the metastasis the administration can be transformed into a format that makes ever more aura in the local environment. For example, the log data may be transferred to a local database allowing idee fixe queries in consideration of be performed against the data.<\p>
