Oracle 1z0-468 問題集

質問 1：
You are using Coherence to store state data that gets read frequently and updated only occasionally. Only 100 or so clients are logged in at any time, and state data is retrieved at login. You are running out of physical memory on the cache server tier, but CPU and network utilization is low. You have 100 GB Solid-State Drives on every system, including clients. You would also like to provide faster access to state data for logged-in users.
Which configuration should you use for your system?
A. Configure Elastic Data on your clients with a near cache.
B. Configure a near cache on your client and a Flash Journal on your cache servers.
C. Because near caching does not work with Elastic Data, configure a near cache and an eviction policy on the cache servers.
D. Configure a near cache on your client and a RAM Journal on your cache servers.
正解：B

質問 2：
Consider an application using a Coherence partitioned cache to store HTTP session data.
There is a load balancer that performs client based affinity (or "sticky load balancing"), meaning that requests that come from a particular user will be routed to the same application server every time. Most user requests will read from the session, thus making near caching topology a good fit.
Which near cache configuration matches the requirements for this application?
A. PRESENT
B. NONE
C. ALL
D. AUTO
正解：A

質問 3：
After going through a sizing exercise, you determine that your application will need 100 storage-enabled JVMs with 4 GB heaps. You have 10 physical machines with 16 cores and 32 GB of RAM.
Which performance-tuning best practice would be violated by deploying your application on these machines?
A. OS socket buffers should be at least 2 M
B. Avoid OS memory swapping/paging by not exhausting physical RAM.
C. Do not exceed one JVM per core.
D. Heap sizes should not be larger than 512 MB.
正解：A

質問 4：
The performance of your query is very slow.
Which two actions would you take to diagnose potential bottlenecks? (Choose two.)
A. Check that indexes are cached against the fields being queried.
B. Check that indexes are registered against the fields being queried.
C. Monitor the task performance time.
D. Use the QueryRecorder to see the query Trace and Explain plan.
E. Monitor the task execution time.
正解：A,E

質問 5：
Your application had experienced an OutOfMemoryError on one of the cache servers. You had configured the JVM to produce a heap dump when OutOfMemoryErrors occur. Your analysis of the heap dump indicates that approximately 90% of the heap is consumed by instances of com.tangosol.util.Binary.
What configuration change will reduce the number of com.tangosol.util.Binary instances?
A. Lower the high-unit threshold on all distributed-scheme backing maps.
B. Increase the size of the thread pool for all cache services.
C. Introduce a near cache for all partitioned caches.
D. Increase the expiry delay on all distributed-scheme backing maps.
正解：B

質問 6：
Which three generational spaces make up the HotStop heap? (Choose three.)
A. Permanent Generation
B. Persistent Generation
C. Old Generation
D. Internal Generation
E. Young Generation
正解：A,C,E

質問 7：
An Oracle RAC Database appliance was initially deployed with two database instances.
The load on the database has now increased and another instance needs to be provisioned.
How would you add one more instance of the earlier deployed appliance?
A. Use the ScaleAppliance operation to scale up the deployed appliance instances from two to three.
B. Add a new single instance database appliance into the previously created assembly and redeploy it.
C. Create and deploy a new assembly containing a single instance database appliance.
D. Use the createAssemblyInstance operation to create a new instance of the existing assembly and deploy the new assembly instance.
正解：D

質問 8：
Identify three situations where Coherence is used (Choose three.)
A. to offload shared services or databases on repeated reads
B. to accelerate complex RDBMS joins by storing data in memory
C. to perform map-reduce aggregations of unstructured data
D. to scale the amount of data accessed on the application or middleware tier
E. to reduce database load by batching and coalescing writes
正解：A,C,E