问题描述
大家好，
我是Azure Analytics的学习者.我知道配置HDInsight群集大约需要15到20分钟.
I am a learner of Azure Analytics. I understand that provisioning the HDInsight cluster will take around 15 to 20 minutes.
如果我的用例是从Azure Blob存储中获取数据并通过以下HDInsight群集执行ELT操作.
If my use case is to take data from Azure Blob storage and do perform ELT operations through following HDInsight clusters.
情况1: (批次)
源->数据工厂-> Azure Blob存储-> Spark-> HBase-> Power BI.
Source -->Data factory-->Azure Blob Storage--> Spark -->HBase -->Power BI .
案例2 :(流)
来源-> Azure事件中心-> Kafka->火花->蜂巢->报告
Source -->Azure Event Hub -->Kafka -->Spark -->Hive --> reports
在以上两种情况下，都必须配置2个或3个HDInsight群集，并且一旦我的操作完成，就将其删除.因此，在日常工作执行中，如果每个集群大约需要20分钟，那么我们如何才能满足该特定工作的SLA 完成?如果我有更多集群来完成数据处理工作量，那么总的配置时间将加起来.在没有统一集群的情况下，客户如何在生产中处理这些方案，而他们仍然符合SLA.
In both the above cases, 2 or 3 HDInsight clusters have to be provisioned and once my operation is completed ,then it would get deleted. So in a daily job execution if each cluster takes around 20 minutes then how can we meet the SLA for that specific job completion? If I have more clusters to complete my data processing workload then the overall provisiong time will get added up. In the absence of unified cluster how these scenarios are handled in production by customers and still they meet SLA.
请澄清我的疑问.谢谢，感谢您的答复
Kindly clarify my doubt. Thanks and appreciate your response
推荐答案
您能否更详细地描述您的SLA?您是否在问如何确保HDI群集设置需要一定的时间?
Could you describe your SLA in more detail? Are you asking how to guarantee a HDI cluster provisioning takes a certain amount of time?
这篇关于日常运营会议SLA上的HDInsight群集配置的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！