Networking Class. Please help me. Suppose the Ethernet transmission algorithm is
ID: 3592022 • Letter: N
Question
Networking Class. Please help me.
Suppose the Ethernet transmission algorithm is modified as follows: After each successful transmission attempt, a host waits one or two slot times before attempting to transmit again and otherwise backs off the usual way.
1. Explain why the capture effect of the previous exercise is now much less likely.
2. Show how the strategy above can now lead to a pair of hosts capturing the Ethernet, alternating transmissions and locking out a third.
3. Propose an alternative approach for example by modifying the exponential backoff. What aspects of a station’s history might be used as parameters to the modified backoff?
Explanation / Answer
Enhanced user access control
This release adds granularity to BIG-IP access control. For any BIG-IP user, a BIG-IP administrator with the appropriate user role can now grant user access to multiple administrative partitions (instead of access to one or all only), and can assign multiple user roles to the user, one for each partition to which the user has access.
Hardware/platform support/maintenance
DSCP mapped to eight hardware CoS queues on egress
This release provides support for traffic being prioritized and dropped selectively using Differentiated Services Code Point (DSCP), based on a limited number of traffic classes that are mapped to eight user-configurable Class of Service (CoS) priorities on egress. This feature is supported only on these platforms: VIPRION B2250 blade, VIPRION B2150 blade, VIPRION B2100 blade, VIPRION B4300 blade, BIG-IP 10000 Series platform, BIG-IP 7000 Series platform, and BIG-IP 5000 Series platform.
Disk erase on SSD and HDD platforms
This release now provides end users with the ability to perform a single pass/zero write disk erase operation of solid-state disk (SSDs) drives and hard disk drives (HDDs). For more information, see SOL15521: Using the 'Security Erase Unit' ATA command to perform a disk erase for SSDs and HDDs, available at support.f5.com/kb/en-us/solutions/public/15000/500/sol15521.html.
BIG-IP 2000/4000 Series appliance platforms L2 enhancements
This release introduces enhanced L2 support, including these features on the BIG-IP 2000/4000 appliance platforms: STP (Spanning Tree Protocol), LLDP (Link Layer Discovery Protocol), ARL (Address Resolution Table), and enhanced Traffic Management Shell (tmsh) Layer 2 Forwarding table commands.
IPv6 support in ePVA (requires 11.6.0 HF5)
The 11.6.0 HF5 release features support for IPv6 in ePVA for the VIPRION B2250 blade.
UC-APL certification
This release includes features that support UC-APL (Unified Capabilities Approved Product List) certification requirements, including: smart card (CAC) authentication to the management interface, configurable banners for confidentiality, FIPS 140-2 compliance, SSL/TLS key requirements, and Appliance mode for DISA/STIG.
System support for multiple hardware (FPGA bitstream) profiles
This release adds system support that enables users to choose from two different available hardware (FPGA firmware) profiles based on provisioning: standard balanced performance profile and an L4-optimized performance profile. This feature is currently available only on the VIPRION B2250 blade in a C2200 chassis.
Traffic group limit increased from 15 to 127
This release supports a maximum of 127 traffic groups that you can configure within a Sync-Failover device group. Earlier releases supported a maximum of 15 traffic groups only.
General Functionality
Object move and rename (early access)
This release provides early access to the feature that enables move/rename of specific BIG-IP object types, such as virtual servers, virtual addresses, pools (implicitly moves pool members), nodes, monitors, profiles, iRules, iApps, device names, self IP addresses, iCall, and folders. Note that this functionality is not provided for VLANs or Partitions.
L7 Policy Matching Enhancements
This release provides a variety of enhancements to L7 policy matching (CPM).
Cubic and Westwood+ congestion control
This release adds the Cubic and Westwood+ congestion control algorithms.
Early retransmit
This release introduces support for an experimental RFC to recover lost segments quickly.
Dynamic TCP tuning
Modification of TCP profile parameters via iRules.
Tail loss probe
This release contains an enhancement to reduce the impact of retransmission timeouts (RTO) on web transactions.
TCP profile redesign
This release provides a redesign of the TCP profile page, to enhance usability.
FIX protocol-based routing together with low latency
FIX data that is available in the first 2144 bytes of a flow can be parsed and used in traffic management, such as routing a flow to a specific backend server based on SenderCompID. Performance L4 virtual servers can extract application data to make the traffic management decisions. Once the decision has been made, the flow is moved to the ePVA for low-latency TCP-based data transfer.
Adaptive response time monitoring
This release provides Adaptive response time monitoring, which measures the amount of time between when the BIG-IP system sends a probe to a resource and when the system receives a response from the resource. It adds an extra dimension to existing monitor capabilities. Use adaptive response time monitoring to enhance server utilization under heavy load and to optimize moderately configurable web applications that are served by servers with limited capacity.
Populate pools by FQDN
This release includes the ability to configure a BIG-IP system with nodes and pool members that are identified with fully-qualified domain names (FQDNs). When configuring pool members with FQDN, addresses dynamically follow DNS changes. Fully dynamic DNS-managed pools may even be created.
Device Trust Group on the Device Management Overview page
The Device Management Overview page in the BIG-IP Configuration utility now reports the status of the special trust device group, which contains all devices in the local trust domain. This new status on the Overview page can help users troubleshoot and resolve sync issues, which are sometimes caused by device trust not being properly established.
IPFIX support with iRules
Ability to use iRules to generate log messages encoded in IPFIX/NetFlow format, containing standard and custom Information Elements.
Support Alert Functionality
The new alert system aligns with and is part of the Unified Logging Infrastructure, with its configuration officially part of standard MCP schemas/CMI/DG, and so on, so that alerts can be raised for any message that originated in the system that is potentially destined for any/all endpoints, including offbox destinations.
Net-SNMP Upgraded to Version 5.7.2
The Net-SNMP software on the BIG-IP system is now upgraded to version 5.7.2.
Kernel Upgraded to RedHat 6.4 Version
The kernel on the BIG-IP system is now upgraded to RedHat 6.4 version 2.6.32-358.23.2.
HTTP 2.0 (experimental) Profile
Local Traffic functionality now includes an HTTP/2 profile type that you can use to manage HTTP/2 traffic, improving the efficiency of network resources while reducing the perceived latency of requests and responses. The Local Traffic HTTP/2 profile enables you to achieve these advantages by multiplexing streams and compressing headers with Transport Layer Security (TLS) or Secure Sockets Layer (SSL) security. Note that subsequent versions of the HTTP/2 protocol might be incompatible with this release. The HTTP 2.0 specification is currently in a draft phase (draft 13).
SPDY 3.1
This release supports SPDY version 3.1 functionality.
iCheck functionality improves monitors scalability
In this release, the BIG-IP system includes new iCheck functionality, which improves scalability of FTP, SMTP, POP3, and IMAP monitors. iCheck functionality supports more monitors, while reducing the load on BIG-IP systems. For example, FTP monitoring provides a 600% improvement in sustained monitor performance. Additionally, iCheck functionality provides smoother performance characteristics as monitors approach full capacity. For example, F5 Networks tested 6,000 monitors showing smooth traffic characteristics throughout the range.
High performance SIP proxy
In this release, you can use the BIG-IP system as a Session Initiation Protocol (SIP) proxy. When the BIG-IP system is placed between your SIP routers, session border controllers, and soft switches, you can configure the system to route and load balance SIP messages across the servers on your SIP network.
IPSec IKEv2 support
IPsec options on Big-IP systems now include support for IKEv2. When you configure IKE peers, you can choose between IKEv1 and IKEv2. If you choose IKEv2, you have the additional benefit of using route domains.
Bandwidth measurement per subscriber and/or flow
This release includes a mechanism for bandwidth measurement (rate or bytes) per subscriber, per application, or per flow. Other elements in the network can use this information to dynamically apply relevant services, for example, video encoding.
vCMP
802.1QinQ on switch enabled platforms
This release supports IEEE 802.1QinQ on switch-enabled platforms: 5000, 7000, 10000, B2100, B2200, B4300 series platforms, which allows for overlapping VLAN IDs, particularly benefiting vCMP or Partitions and route-domain deployments
vCMP virtual disk templates
You can speed up vCMP guest deployment by using virtual disk templates. On the first vCMP guest installation, the vCMP system creates a virtual disk template for the specific initial-image (and, if present, initial-hotfix). For subsequent guest installations of the same initial-image/hotfix, vCMP host administrators can use the virtual disk template, which speeds up the deployment process.
vCMP guest data visible from host
This version provides a summary of vCMP guest data from within the vCMP host. This data facilitates access to the current state of each guest. The BIG-IP Configuration utility shows the current active software image, the provisioned modules, and HA status. You can access each guest for additional information, such as Installed Images and Available Images, and other Resource Provisioned information such as License Status, Required Disk and Required Memory. The Guest name link opens the guest's Properties tab. The HA Status shows a Failed link when there is an HA Status failure. The link opens the guest's HA Status tab.
vCMP guest access to ISO/hotfix images from hypervisor
vCMP now allows file access from within a vCMP guest to the images stored on the host (hypervisor) side. This facilitates installation of ISO/hotfix images to guests and reduces storage space for those images.
Historical vCMP Statistics
From the vCMP host, you can view detailed historical vCMP statistics in the Analytics section of the Configuration utility. The statistics provide an overview of vCMP performance, network throughput, CPU usage, and disk usage in graphical form. You can customize the information that is displayed, the time periods, and what information you want to appear on the overview screen.
CGNAT
CGNAT: Improved compatibility of CGNAT ALG port selection
This release improves compatibility of application layer gateway (ALG) profiles with Carrier-Grade NAT (CGNAT) port picking methods, including Deterministic NAT (DNAT) and Port Block Allocation (PBA) translation modes. Improved CGNAT port selection compliance is now available in the FTP, SIP and RTSP ALG profiles, allowing the ALGs to select the correct port based on the subscriber, and provide reliable reverse mappings of translated addresses for all traffic.
CGNAT: Port Block Allocation support
Port block allocation (PBA) mode is an address-port translation mode option that reduces CGNAT logging, by logging only the allocation and release of each block of ports. When a subscriber first establishes a network connection, the BIG-IP system reserves a block of ports on a single IP address for that subscriber, and logs the block allocation. The system releases the block when no more connections are using it, and logs the block release. This functionality reduces the logging overhead, which significantly decreases log storage and improves system performance when compared to NAPT logging, because the CGNAT logs only the allocation and release of each block of ports once.