Closed Questions Flashcards
In a three-layer network architecture of a Datacenter, which is the layer that is typically associated to the TOR switch?
A - Access
B - Core
C - Aggregation
D - None of the others
A
Which sentence about Tier Level 4 of datacenter is not correct?
A - Fault-tolerant site infrastructure with availability of 99.995%
B - Meets or exceeds all Tier 3 requirements
C - Single non-redundant distribution path serving the IT equipment
D - All cooling equipment is independently dual-powere
C
Which is the main characteristic of the D-Cell topology for data center network architectures?
A - the network is organized in a recursive way
B - The network comprises edge, aggregation and core layers
C - the network is organized in a hierarchical way
D - The network is organized in a distributed way
A
What is a hybrid cloud computing system?
A - a cloud used for a single organization
B - a cloud whose energy is produced both by diesel and batteries
C - a cloud shared by several organization
D - a cloud composed by private and public cloud systems
D
Paravirtualization cannot be used with unmodified OS
True or False
True
Servers in data centers are like regular personal-computers with the same form factor
True or False
False
Warehouse-scale computers run a large number of very small applications
True or False
False
In datacenter storage context, DAS means Distributed Application Storage
True or False
False (Direct Attached Storage)
An Edge Computing system provides higher performance (in terms of memory and computation) than Internet-of-Things devices
True or False
True
What is a Network Attached Storage (NAS)?
A - a storage system directly attached to a server or workstation
B - a computer connected to a network that provides only file-based data storage services to other devices
C - a remote storage unit connected to PC using a specific networking technology
D - a computer connected to a network that provides computation to other devices
B
Downtime is a period of time, or a percentage of a time span, when a system is unavailable or offline __________
A - none of the other
B - because of routine maintenance
C - because of system crashes
D - regardless of the cause
D
In a three-layer network architecture of a Datacenter, which is the layer that is typically associated to the TOR switch?
A - Access
B - Core
C - Aggregation
D - None of the others
A
Which definition is not encompassed by Dependability:
A - Reversibility: ability to reverse a broken service
B - Reliability: continuity of correct service
C - Maintainability: reparation to restore correct service
D - Availability: readiness for correct service
A
Racks shelves may contain switches
True or False
True
In the context of SSDs, Blocks are the smallest unit that can be read/written
True or False
False
In Solid State Drives (SSDs), the smallest units that can be written or read are typically referred to as “pages,” not blocks. However, SSDs manage data internally in blocks, which consist of multiple pages. So, while pages are the smallest units that can be written or read, blocks are important for the internal management of data within an SSD.
Paravirtualization is a particular type of kernel-level Virtualization that does not rely on a host OS
True or False
False
The structure (or topology) of a reliability block diagram is always the same as the structure (or topology) of the modeled system
True or False
False
RBD may not always precisely match the physical or logical structure of the modeled system. Sometimes, for the sake of analysis or simplicity, the RBD may abstract or combine certain components or subsystems. Additionally, the RBD may include redundancy or mitigation strategies that are not explicitly present in the physical system but are included for reliability analysis purposes.
A system virtual machine provides a complete system environment that can support an operating system (potentially with many user processes)
True or False
True
Edge computing can improve the performance and reduce the latency of applica- tions by reducing the distance data has to travel.
True or False
True
The Internet of Things (IoT) layer is part of the computing continuum.
True or False
True
NASs can provide high-speed, low-latency access to storage resources for applica- tions that require fast access to data.
True or False
False
Cooling towers use cold water to remove heat and to function effectively.
True or Flase
False
A PUE of 1.0 means that all of the power consumed by a data center is being used by the IT equipment.
True or False
True
DAS devices are typically shared by multiple servers.
True or False
False
FPGAs are more power-efficient and offer higher performance compared to CPUs for certain types of workloads
True or False
True
Full virtualization provides better performance than paravirtualization.
True or False
False
Closed loop cooling systems are not suitable for high-density datacenters where heat dissipation is critical.
True or False
False
Cloud architectures are designed to provide on-demand access to shared computing resources.
True or False
True
Hot/cold aisle containment reduces the efficiency of the cooling method for data-centers
True or False
False
Edge computing is not a suitable alternative to cloud computing for applications that require large-scale data processing.
True or False
True
RAID 0 provides the best read and write performance of any RAID level composed of the same amount of disks.
True or False
True
Datacenters within the same Geographic Areas share the same power and network infrastructure
True or False
False
TOR switches are typically used in small-scale datacenter deployments and are not suitable for larger networks.
True or False
False
Virtualization can help reduce energy consumption by consolidating multiple physical machines into fewer physical ones.
True or False
True
POD, Virtual-Chassis, DCell and BCube models are all evolutions of the leaf-spine network architectures.
True or False
False
Type 2 hypervisors run on top of a guest operating system that provides indirect access to hardware resources.
True or False
False
In-row cooling systems allow for higher server densities and increased rack power densities in datacenters.
True or False
True
DAS (Direct Attached Storage) is a storage system directly attached to the datacenter network.
True or False
False
The computing continuum refers to the idea that computing devices exist on a spectrum, from small, low-power devices to large, high-performance machines.
True or false
True
Virtualization can improve disaster recovery by enabling quick and easy backup
and restoration of virtual machines.
True
In-row cooling systems are less efficient at removing heat compared to traditional perimeter cooling systems.
False
In-Row Cooling
- Placement: In-row cooling units are placed between server racks, directly within the rows of equipment.
- Cooling Efficiency: These units provide cooling closer to the heat source (i.e., the servers), which can lead to more efficient and targeted cooling. The cold air is directed directly to the servers, and hot air is removed from the area more quickly.
Perimeter Cooling
- Placement: Perimeter cooling units are typically placed along the walls or at the perimeter of the data center.
- Cooling Efficiency: These units rely on distributing cold air through the raised floor or through ducts that deliver cold air to the front of the server racks. The return air is then collected and cooled down again. This method can be less efficient due to potential mixing of hot and cold air.
Oversubscription is not used in TOR switches
False
Oversubscription refers to a design strategy where the aggregate bandwidth of the uplinks (the links connecting to higher-tier switches or routers) is less than the aggregate bandwidth of the downlinks (the links connecting to end devices or lower-tier switches). This means that the network is designed with the expectation that not all devices will need to use their maximum bandwidth simultaneously.
Liquid cooling is a new technology that is not yet suitable for use in productio data centers
False
NAS devices can be used to share files across multiple clients, providing a centralized storage location for an organization.
True
RAID 10 can achieve better sequential read performance than RAID 0.
False
Multiple Compute Regions permit not only disaster recovery but also a very efficient application-level synchronous replication
False
Cloud architectures can help organizations to reduce their carbon footprint by enabling resource sharing and consolidation.
True
FPGAs are more power-efficient and offer higher performance compared to CPUs for certain types of workloads
True
GPUs (Graphics Processing Units) are primarily used in datacenters for accelerating graphical applications and gaming.
True
SANs typically use Fibre Channel or iSCSI to connect servers to storage devices.
True
• Fibre Channel (FC): High-performance, low-latency, and highly reliable, ideal for enterprise-grade SANs.
• iSCSI: Cost-effective and easy to implement using existing IP networks, suitable for small to medium-sized deployments.
• Fibre Channel over Ethernet (FCoE): Combines the benefits of FC and Ethernet, simplifying infrastructure and reducing costs.
• InfiniBand: Extremely high performance, used in specialized environments requiring very low latency and high throughput.
GPUs are generally less expensive to deploy and maintain compared to CPUs in datacenter environments.
False
Liquid immersion cooling can damage servers if not done properly.
True
TPUs are widely available and can be easily procured for use in most datacenter environments
True
Datacenter tiers are a standardized classification system used to categorize the reliability and availability of a data center.
True
Hypervisors are only used in cloud computing environments.
False
Cloud architectures rely on service-level agreements (SLAs) to ensure uptime and availability
False
Open loop (free) cooling refers to the use of cold outside air to either help produce chilled water or directly cool servers.
True
RAID 5 can handle sequential reads and writes better than RAID 10 with the same amount of disks
True
Monolithic hypervisors are better suited for high-performance computing than microkernel hypervisors.
True
A PUE of 1.0 means that all of the power consumed by a data center is being used by the IT equipment.
True
Yes, that’s correct. Power Usage Effectiveness (PUE) is a metric used to determine the energy efficiency of a data center. It is calculated as the ratio of the total amount of energy used by the data center to the energy used by the IT equipment alone. The formula for PUE is:
[ \text{PUE} = \frac{\text{Total Facility Energy}}{\text{IT Equipment Energy}} ]
Where:
- Total Facility Energy includes all the energy consumed by the data center, encompassing everything from cooling systems, lighting, power distribution losses, and other non-IT infrastructure.
- IT Equipment Energy is the energy consumed by servers, storage, networking devices, and other IT equipment.
- PUE of 1.0: This indicates perfect energy efficiency, meaning that all the power consumed by the data center is being used solely by the IT equipment, with no energy wasted on cooling, power conversion, lighting, or any other overhead. This is an ideal scenario and practically very difficult to achieve.
- PUE greater than 1.0: Indicates that some of the power is being consumed by non-IT infrastructure. For instance, a PUE of 1.5 means that for every 1 watt of power consumed by IT equipment, an additional 0.5 watts are consumed by the data center’s overhead (cooling, power conversion, etc.).
GPUs require specialized software and programming frameworks to fully leverage their parallel processing capabiliti
True
Warehouse-scale computers require specialized cooling and power management systems
True
Hot/cold aisle containment reduces the efficiency of the cooling method for data-center
False
SSDs use spinning disks to store data.
False
In a leaf-spine topology, adding or removing a leaf switch does not impact the connectivity of other switches in the network.
True
POD, Virtual-Chassis, DCell and BCube models are all evolutions of the leaf-spine network architectures.
False