Professional Documents
Culture Documents
AWS S3 Interview Questions
AWS S3 Interview Questions
Below are the list of Best AWS S3 interview questions and Answers
Amazon S3 (Simple Storage Service) is a simple web service interface that allows huge amount of data
storage and retrival from anywhere from the internet.Its provides developers highly scalable, reliable,
fast and low cost data storage infrastructure.
In S3 bucket you can store unlimited volume of data and number of objects. A single Amazon S3 objects
can be a size of range 0 bytes to 5 terabytes. In single upload request You can put an object of around 5
GB but you must have to enable Multipart Upload capability.
use aws s3 ls --recursive command on aws CLI to list all files or objects under the specified directory or
prefix
After that, if you are using Unix or Linux system run below command to install AWS CLI
You can use invalidations to clear AWS S3 cache. To create invalidations login to AWS Console and go
to Distribution Settings > Invalidations > Create Invalidation. Once invalidation is created just type path
of file or a wildcard to purge the cache.
9) What are options available for protecting data at rest in Amazon S3?
Versioning allows us to keep multiple variants of an object in a bucket. Versioning helps us to restore an
object to a previous or specific version of an object. You can take advantage of versioning to recover a
deleted or mistakenly overwritten object.
Versioning helps you to keep multiple versions of an object in one bucket. Here are simple steps to
enable versioning on an S3 bucket.
It is a FUSE filesystem. Amazon web services simple storage service supports it. It can be operated with
two different methods
1.Command method-
In this type of mode, s3fs is eligible for managing Amazon s3 buckets in several efficient methods.
2. Mount method-
4. Hadoop clusters running on Amazon EMR use ______ instances as virtual Linux servers for the master
and slave nodes.
a)EC2
b)EC3
c)EC4
d)None of the mentioned
View Answer
Answer: a
Explanation: Amazon EMR has made enhancements to Hadoop and other open-source applications to
work seamlessly with AWS.
6. Amazon EMR uses Hadoop processing combined with several __________ products.
a) AWS
b) ASQ
c) AMR
d) AWES
View Answer
Answer: a
Explanation: Amazon Elastic MapReduce (Amazon EMR) is a web service that makes it easy to process
large amounts of data efficiently
7. ___________ is an RPC framework that defines a compact binary serialization format used to persist
data structures for later analysis.
a) Pig
b) Hive
c) Thrift
d) None of the mentioned
View Answer
Answer: b
Explanation: Amazon EMR does not support Hive Authorization.
Q.2 What is the function of the command- ec2-create-group Create Security Group?
Ans. Creates a new security group for use with your account.
Q.4 When will you incur costs with an Elastic IP address (EIP)?
Ans. When it allocates and associate with a stopped instance.
Q.8 What kind of network performance parameters can you expect when you launch instances in
cluster placement group?
Ans. The network performance differs between the instance type and network performance
specification, in a placement group the user can expect till
10 Gbps in a single-flow.
20 Gbps in multi-flow for an instance full duplex.
Network traffic excluding the selected group will limit to 5 Gbps(full duplex).
Q.9 Is one Elastic IP address enough for each instance that I actually have running?
Ans. Each instance comes with its own personal and public address. The personal address is associated
solely with the instance and it came back to Amazon EC2 only if it’s stopped or terminated. Similarly, the
public address is associated solely with the instance till it’s stopped or terminated. However, this will get
replaced by the Elastic information processing address that stays with the instance as long because the
user doesn’t manually detach it. However, what if you’re hosting multiple websites on your EC2 server,
therein case you’ll need quite one Elastic IP address.
Q.10 You need to set up an Amazon S3 bucket to serve static assets for your public-facing internet
application. That technique can make sure that all objects uploaded to the bucket are set to public
read?
Ans. Configure the bucket policy to set all objects to public read. Making changes to every object is not
preferred as it’s better to set the policy for the complete bucket. IAM is used to give more granular
permissions since this is a website, all objects would be public by default.
Q.14 Can a user connect the company’s data centre to the Amazon Cloud?
Ans. Yes, you’ll be able to try this by establishing a VPN(Virtual Private Network) affiliation between your
company’s network and your VPC (Virtual personal Cloud), this may permit you to act together with
your EC2 instances as if they were inside your existing network.
Q.15 Is it possible to alter the personal IP addresses of associate EC2 whereas it’s running/stopped in
an exceedingly VPC?
Ans. The primary personal IP address connects with the instance throughout its lifespan and can’t
modify, but secondary personal addresses will unassigned, appointed or rapt between interfaces or
instances at any purpose.
Q.17 If a user uses Amazon CloudFront, is able to use direct hook up with transfer objects from his
data centre?
Ans. Yes. Amazon CloudFront supports custom origins. It also includes origins from outside of AWS.
With AWS Direct Connect, the user will charge with the different information transfer rates.
Q.18 If AWS Direct Connect fails, can the user loose connectivity?
Ans. If a backup AWS Direct Connect design, within the event of a failure it’ll change by reversal to the
second. It suggests changing two-way Forwarding Detection (BFD) once configuring your connections to
make sure quicker detection and failover. On the opposite hand, if you’ve got design a backup IPsec VPN
affiliation instead, all VPC traffic can failover to the backup VPN affiliation mechanically. Traffic to/from
public resources like Amazon S3 is going to route over the net. If a user doesn’t have a
backup AWS Direct Connect link or an IPsec VPN link, then Amazon VPC traffic will lead to a rise in the
event of a failure.
Q.20 When should a user prefer Provisioned IOPS over normal RDS storage?
Ans. If you’ve got batch-oriented workloads. Provisioned IOPS deliver high IO rates however on the
opposite hand it’s high-ticket still. Execution workloads are not in need of manual intervention as they
allow full utilization of systems, which states that provisioned IOPS are most popular for batch-oriented
work.
Q.21 How are Amazon RDS, DynamoDB and Redshift different from each other?
Ans. Amazon RDS may a direct service for relative databases; it manages fixing, upgrading, backing of
information etc. of databases for you while not your intervention. RDS is probably a database
management service for structure data solely.
Amazon DynamoDB, on the opposite hand, maybe a NoSQL info service, NoSQL deals with unstructured
information.
Amazon Redshift, is a completely different service, its data warehouse product and employ in data
analysis.
Q.22 Can a user run more than one DB instance for Amazon RDS for free?
Ans. Yes. A user can run over one Single-AZ small info instance, that too for free! but, use exceeding 750
instance hours, across all Amazon RDS Single-AZ small db instances, across all eligible database engines
and regions, will lead to the billing at normal Amazon RDS costs. For example: if you run 2 Single-AZ
small db instances for four hundred hours every in an exceedingly single month, you may accumulate
800 instance hours of usage, of that 750 hours are free. You may bill for the remaining fifty hours at the
quality Amazon RDS value.
Q.23 Can a user retrieve only a particular part of the information if he has a nested JSON data in
DynamoDB?
Ans. Yes. Once using the GetItem, BatchGetItem, question or Scan Apis, the user will able to outline a
Projection Expression to see that attributes ought to retrieve from the table. Those attributes will
embody scalars, sets, or parts of a JSON document.
Q.24 A corporation is deploying a brand new two-tier internet application in AWS. The corporate has
restricted workers and needs high handiness, and also the application needs advanced queries and
table joins. That configuration provides the answer for the company’s requirements?
Ans. Amazon DynamoDB. DynamoDB has the flexibility to scale over RDS or the other computer
database service, so DynamoDB would the apt selection.
Q.25 What will cause to the backups and dB Snapshots if a user deletes the dB Instance?
Ans. When you delete a DB instance, you have got an option of making a final dB snapshot, if you are
doing that you simply will restore your info from that photograph. AWS RDS retains these user-created
dB snapshots together with all different manually created dB snapshots when the instance delete. In
addition, automated backups delete and solely manually created dB Snapshots preserve.
Q.4 How can you modify the instance kind for instances that are running in your application tier
and are using Auto Scaling. Where can you modify it from the subsequent areas?
Ans. Auto Scaling launch configuration. Amazon Auto-scaling tags configuration is employed to connect
metadata to your instances, to vary the instance kind you’ve got to use auto-scaling launch
configuration.
Follow this link to know more about AWS Autoscaling
Q.5 The user got a content management system running on an Amazon EC2 instance that’s
approaching 100% processor utilization. Can that possibility scale backload on the Amazon EC2
instance?
Ans. Produces a load balancer, and register the Amazon EC2 instance with it. Making alone an auto-
scaling cluster won’t solve the difficulty, till you attach a load balancer to it. Once you attach a load
balancer to associate autoscaling cluster, it’ll with efficiency distribute the load among all the instances.
Possibility B – CloudFront could be a CDN, it’s a data transfer tool thus won’t facilitate scale backload on
the EC2 instance. Equally the opposite possibility – Launch configuration could be a templet for a
configuration that has no reference to reducing masses.
Q.6 Once ought to use a Classic Load Balancer associate once to use an Application load balancer?
Ans. A Classic Load Balancer is good for easy load levelling of traffic across multiple EC2 instances. The
associate Application Load Balancer is good for microservices or container-based architectures.
Wherever there’s a desire to route traffic to multiple services or load balance across multiple ports on
an identical EC2 instance.
Q.8 Once an instance is unhealthy, it’s terminated and replaced with a replacement one, that of the
subsequent.
Ans. Once ELB detects that associate instance is unhealthy, it starts routing incoming traffic to
alternative healthy instances within the region. If all the instances in an exceeding region become
unhealthy, and if you’ve got instances in another handiness zone/region, your traffic is directed to them.
When the instances get healthy they’re re-routed back to the first instances.
Best Amazon Interview Questions and Answers. Q- 2,5,8
Best Amazon Interview Questions and Answers. Q- 1,3,4,6,7
Q.10 The application of the should retrieve data from your user’s mobile every five minutes and also
the data keep in DynamoDB, later on, a daily basis at a specific time the data is extracted into S3 on a
per-user basis then your application is later wont to visualize the data to the user. The user is asked to
optimize the design of the backend system to lower price, what would you recommend?
Ans. Introduce Amazon Elasticache to cache scans from the Amazon DynamoDB table and scale back
provisioned read throughput. Since our work needs the data to be extracted and analyzed, to optimize
this method someone would use provisioned IO, however since it’s costly, using an ElastiCache memory
instead to cache the ends up in the memory will scale back the provisioned browse outturn and thence
cut back price while not affecting the performance.
Q.11 The user is running an internet site on EC2 instances deployed across multiple handiness Zones
with a Multi-AZ RDS MySQL extra large db Instance. The location performs a high variety of tiny reads
and writes per second and depends on an ultimate consistency model. When comprehensive tests you
discover that there’s browse competition on RDS MySQL. Justify any pair of suitable needs.
Ans. Deploy ElastiCache in-memory cache running in every handiness zone
Increase the RDS MySQL Instance size and Implement provisioned IOP.
Since it will plenty have read writes, provisioned IO could become costly. However, we’d like high
performance moreover, thus the data is cached using ElastiCache which might be used for ofttimes
reading the information. As for RDS since browse competition is occurring, the instance size ought to be
redoubled and provisioned IO ought to be introduced to extend the performance.
Follow this link to know about AWS Relational Database
Q.12 A startup is running a pilot preparation of around a hundred sensors to live street noise and air
quality in urban areas for three months. it absolutely was noted that each month around 4GB of
device data is generated. The corporate uses a load balanced auto-scaled layer of EC2 instances and
RDS info with five hundred GB standard storage. The pilot was a hit and currently, they need to
Q.13 What happens if CloudTrail is turned on for the account. However, the Amazon S3 bucket isn’t
organized with the proper policy?
Ans. CloudTrail files are delivered consistent with S3 bucket policies. If the bucket isn’t organized or is
misconfigured, CloudTrail may not be ready to deliver the log files.
Q.14 How can a user transfer the existing name registration to Amazon Route fifty-three while not
disrupting the existing internet traffic?
Ans. You will get to get an inventory of the DNS record data for your name initial, it’s typically out there
within the kind of a “zone file” that you just will get from your existing DNS supplier. Once you receive
the DNS record data, you’ll be able to use Route 53’s Management Console or easy web-services
interface to form a hosted zone which will store your DNS records for your name and follow its transfer
method. It conjointly includes steps like changing the name servers for your name to those related to
your hosted zone. For finishing the method you’ve got to contact the registrar with whom you
registered your name and follow the transfer process. As long as your registrar propagates the new
name server delegations, your DNS queries can begin to urge answered.
Follow this link to explore about AWS Management Console
Q.17 How is AWS Elastic stalk completely different than AWS OpsWorks?
Ans. AWS Elastic stalk could be an application management platform whereas OpsWorks is a
configuration management platform. The stalk is a simple to use service that is employed for deploying
and scaling internet applications developed with Java, Net, PHP, Node.js, Python Programming, Ruby,
Go and labourer. Customers transfer their code and Elastic stalk automatically handles the preparation.
The applying is going to be able to use with none infrastructure or resource configuration.
Read more about AWS Elastic Beanstalk (AWS EBS)
Q.23 Make a case for Stopping, Starting, and Terminating an Amazon Ec2 Instance?
Ans. Stopping and beginning an instance: Once an instance is stopped, the instance performs a
traditional closing so transitions to a stopped state. All of its Amazon east by south volumes stay
connected, and you’ll be able to begin the instance once more at a later time. You’re not charged for
extra instance hours whereas the instance is during a stopped state.
Terminating an instance: Once an instance is terminated, the instance performs a traditional closing,
and then the connected Amazon EBS volumes are deleted unless the volume’s deleteOnTermination
attribute is ready to false. The instance itself is additionally deleted, and you can’t begin the instance
once more at a later time.
Q.24 Make a case for Elastic Block Storage? What kind of Performance are you able to Expect? How
does one Back It Up? How does one Improve Performance?
Ans. EBS could be a virtualized SAN or enclosure network. Which means its RAID storage to start out
with, therefore it’s redundant and fault tolerant. If disks die therein RAID you don’t lose data. Great! It’s
additionally virtualized, therefore you’ll be able to provision and apportion storage, and connect it to
your server with varied API calls. No occupation the storage skilled and asking him or her to run
specialized commands from the hardware marketer.
Performance on EBS will exhibit variability. That is, it will go higher than the SLA performance level, and
then drop below it. The SLA provides you with a mean disk I/O rate you’ll be able to expect. This may
frustrate some people, particularly performance consultants UN agency expect reliable and consistent
disk throughout on a server. Ancient physically hosted servers behave that means. Virtual AWS instances
don’t.
Q.1 The user wants to launch a Linux instance in AWS EC2. While configuring security group, the user
has selected SSH, HTTP, HTTPS protocol. What is the need to select SSH?
Answer - To verify that there is a rule that allows traffic from EC2 Instance to The user computer.
Read More about AWS EC2 Tutorial (Amazon Elastic Compute Cloud)
Q.2 The user needs to quickly set up an email service because a client needs to start using it in the
next hour. Which one can be the logical choice as there are several options available to set it up?
What can be the best option to set up AWS service would best meet the needs of the client?
Answer - Amazon SES console
Q.4 Load Balancer and DNS service come under which type of cloud service?
Answer- IAAS-Storage.
Q.5 The user has an EC2 instance that has an unencrypted volume. The user wants to create another
encrypted volume from this unencrypted volume. Which of the following steps can achieve this?
Answer- Create a snapshot of the unencrypted volume (applying encryption parameters). Copy the
snapshot and create a volume from the copied snapshot.
Q.6 Where does the user specify the maximum number of instances with the auto scaling commands?
Answer- Auto scaling Launch Config.
Q.7 A user identifies that a huge data download is occurring on his instance. He has already set the
auto scaling policy to increase the instance count. When the network Input Output increase beyond a
threshold limits how can the user ensure that this temporary event does not result in scaling?
Answer- Suspend scaling.
Following are some commonly asked Amazon Web Service Interview Questions –
Q.15 How will a user manage firewall settings for the instance?
Answer- Yes. The user will manage the data traffic for the instances by using the Lightsail firewall. From
the Lightsail console, the user will set rules regarding that ports of the user instance are publically
accessible for various kinds of traffic.
Let’s have a look at the latest and most asked Amazon Web Service Interview Questions –
Follow this link to know about AWS VPC (Virtual Private Cloud)
Q.18 What are Availability Zones?
Answer- Availability Zones are collections of knowledge centres that run on physically distinct, freelance
infrastructure and are built to be extremely reliable. Common points of failure like generators and
cooling instrumentality aren’t shared between convenience Zones. In addition, convenience Zones are
physically separate, in order that even extraordinarily uncommon disasters like fires, tornados, or
flooding. This will have an effect on solely one convenience Zone.
Q.21 When would a user use Amazon RDS vs. Amazon EC2 relational database AMIs?
Answer- Amazon internet Services provides a variety of info alternatives for developers. Amazon RDS
allows the user to run a completely featured on-line electronic information service whereas offloading
database administration. Using several on-line databases AMIs on Amazon EC2 permits the user to
manage the user’s own on-line database within the cloud. There are necessary variations between these
alternatives which will build an added acceptable for the user use case. See Cloud Databases with AWS
for steering on that answer is best for the user.
Q.25 How do I specify that supported db engine version I might like the db instance to run?
Answer- The user can specify any presently supported version (major and minor) once making a
replacement db instance via the Launch DB Instance operation within the AWS Management Console or
the Create DB Instance API. Please note that not each info engine version is out there in each AWS
region.
Q.1 How do I specify that supported DB engine version I would like the DB instance to run?
Answer- The user can specify any supported version which is available at present (major and minor)
once making a replacement DB instance via the Launch DB Instance operation within the AWS
Management Console or the Create DB Instance API. Please note that not each info engine version is
out there in each AWS region.
Q.2 How are reserved instances totally different from on-demand dB instances?
Answer- Functionally, reserved instances and on-demand dB instances are precisely the same. The sole
distinction is how the user dB instance(s) are billed: With Reserved Instances. The user get a 1 or 3-year
reservation and reciprocally to receive a lower effective hourly usage rate for the period of the term.
Unless the user gets reserve instances during a Region. All dB instances are going to beak at on-demand
hourly rates.
Do you know what is Amazon Dynamo DB?
Q.6 What if I even have associate existing dB instance that I’d wish to cover with a reserved instance?
Answer- Simply purchase a dB instance reservation with a similar dB instance category, DB engine,
Multi-AZ possibility and License Model at intervals a similar region because the dB instance the user is
presently running and would really like to order. If the reservation purchase is thriving, Amazon RDS can
automatically apply the user new hourly usage charge to the user existing dB instance.
Q.7 If I sign in for a reserved instance when will the term begin? What happens to the dB instance
when the term ends?
Answer- Pricing changes related to a reserved instance are activated once the user request is received
whereas the payment authorization is processed. The user will be able to follow the standing of the user
reservation on the AWS Account Activity page or by mistreatment the DescribeReservedDBInstances API
or describe-reserved-db-instances command. If the one-time payment cannot with success approve by
ensuing asking amount, the discounted worth won’t become.
When the user reservation term expires, the user reserved instance can revert to the acceptable On-
Demand hourly usage rate for the user dB instance category and Region.
Q.8 How do I manage that dB instances billed at the reserved instance rate?
Answer- The Amazon RDS operations for making, modifying, and deleting dB instances don’t distinguish
between on-Demand and reserved instances. Once computing the user bill, our system can
automatically apply the user Reservation(s) such all eligible dB instances are charged at the lower hourly
reserved dB instance rate.
2. Senerio-Based AWS Interview Questions and Answers
Q.9 How can a user move the existing DB instances outside VPC into the VPC?
Answer- If the user DB instance isn’t in a very VPC, the user will be able to use the AWS Management
Console to simply move the user dB instance into a VPC. See the Amazon RDS User Guide for additional
details. The user will be able to conjointly take a photo of the user DB Instance outside VPC and restore
it to VPC by specifying the db Subnet cluster the user would like to use. Or else, the user will be able to
perform a “Restore to purpose in Time” operation likewise.
Read more about AWS VPC (Virtual Private Cloud)
Q.10 How will a person move the existing dB instances from within VPC to outside VPC?
Answer- Migration of dB Instances from within to outside VPC isn’t supported. For security reasons, a dB
photo of a dB Instance within VPC can’t repair to outside VPC. A similar is true with “Restore to purpose
in Time” practicality.
Q.12 How can a user modify the dB Subnet cluster of the dB Instance?
Answer- An existing dB Subnet cluster will update to feature additional subnets, either for existing
handiness Zones or for brand new handiness Zones additional since the creation of the dB Instance.
Removing subnets from associate existing dB Subnet cluster will cause inconvenience for instances if
they’re running in a very explicit AZ that gets aloof from the subnet cluster. Read the Amazon RDS User
Guide for additional info.
Q.18 Can all the instances in my RDS account sample metrics at an equivalent granularity?
Answer- No. you’ll set totally different granularities for every dB Instance in your RDS account. You’ll
additionally opt for the instances on that you would like to modify enhanced monitoring moreover as
modify the granularity of any instance whenever you would like.
Q.20 How am I able to visualize the metrics generated by RDS enhanced monitoring in CloudWatch?
Answer- The metrics from RDS enhanced monitoring deliver into your CloudWatch Logs account. You’ll
produce metrics filters in CloudWatch from CloudWatch Logs and show the graphs on the CloudWatch
dashboard. For a lot of details, please visit the Amazon CloudWatch page.
Q.21 When should I use CloudWatch rather than the RDS console dashboard?
Answer- You should use CloudWatch if you would like to look at historical data on the far side what’s
accessible on the RDS console dashboard. You’ll monitor your RDS instances in CloudWatch to diagnose
the health of your entire AWS stack in a very single location. Currently, CloudWatch supports
granularities of up to one minute. The values are going to average out for granularities but that.
Q.22 How can a user set up alarms and notifications based on specific metrics?
Answer- Yes. You’ll produce an alarm in CloudWatch that sends a notification once the alarm changes
state. The alarm watches one metric over a period of time that you simply specify. It performs one or a
lot of actions supported the worth of the metric relative to the required threshold over the variety of
your time periods. For a lot of details on CloudWatch alarms, please visit the Amazon CloudWatch
Developer Guide.
Q.23 How do I integrate enhanced monitoring with my tool that I presently use?
Answer- RDS enhance monitoring provides a collection of metrics shaped as JSON payloads that deliver
into your CloudWatch Logs account. The JSON payloads deliver at the coarseness last designed for the
RDS instance.
There are 2 ways that you’ll consume the metrics via a third-party dashboard or application. Monitoring
tools will use CloudWatch Logs Subscriptions to line up a close to real-time feed for the metrics. As an
alternative, you’ll use filters in CloudWatch Logs to bridge metrics across to CloudWatch to and
integrate your application with CloudWatch. Please visit Amazon CloudWatch Documentation for a lot of
details.
Q.25 How can a user scale the DB instance beyond the largest DB instance class and maximum storage
capacity?
Answer- There are many DB instances and classes which supports by the Amazon RDS to meet the
different requirements of the applications. For spreading the data across multiple DB instances the user
can implement partitioning. This is done when the application needs more computing resources than
the largest DB instance class or more storage than the maximum allocation.