Methods for analyzing and load balancing based on server health and devices thereof

Abstract

A method, non-transitory computer readable medium, and health analysis apparatus that monitors network traffic exchanged with a plurality of server devices in a server pool to obtain signal data regarding a plurality of signals associated with the network traffic. A determination is made when there is a sever health anomaly for one or more of the server devices based on an application of a server health prediction model to the signal data. The server health prediction model includes a plurality of predictive health targets each based at least in part on historical signal data for one or more of the signals and having an associated threshold value. A mitigation action is initiated when the determining indicates there is a sever health anomaly for one or more of the server devices.

Description

FIELD

This technology generally relates to computer networks and, more particular, to methods and devices for analyzing server health.

BACKGROUND

Server health is often analyzed based on thresholds established by network administrators corresponding to latency observed in network traffic. For example, network traffic exchanged between server devices and traffic management computing devices that are disposed between client devices and these server devices in a network. However, latency is not a robust or effective metric with respect to server health.

In particular, current methods for analyzing server health based on latency do not facilitate effective load balancing decisions or early recognition of impending server saturation or other server health problems. Additionally, the thresholds used by current server health analysis methods are static and may not accurately identify server health problems, particularly over an extended period of time.

SUMMARY

A method for analyzing server health includes monitoring, by a health analysis apparatus, network traffic exchanged with a plurality of server devices in a server pool to obtain signal data regarding a plurality of signals associated with the network traffic. A determination is made, by the health analysis apparatus, when there is a sever health anomaly for one or more of the server devices based on an application of a server health prediction model to the signal data. The server health prediction model includes a plurality of predictive health targets each based at least in part on historical signal data for one or more of the signals and having an associated threshold value. A mitigation action is initiated, by the health analysis apparatus, when the determining indicates there is a sever health anomaly for one or more of the server devices.

A health analysis apparatus includes memory comprising programmed instructions stored in the memory and one or more processors configured to be capable of executing the programmed instructions stored in the memory to monitor network traffic exchanged with a plurality of server devices in a server pool to obtain signal data regarding a plurality of signals associated with the network traffic. A determination is made when there is a sever health anomaly for one or more of the server devices based on an application of a server health prediction model to the signal data. The server health prediction model includes a plurality of predictive health targets each based at least in part on historical signal data for one or more of the signals and having an associated threshold value. A mitigation action is initiated when the determining indicates there is a sever health anomaly for one or more of the server devices.

A non-transitory computer readable medium having stored thereon instructions for analyzing server health traffic includes executable code which when executed by one or more processors, causes the one or more processors to perform steps including monitoring network traffic exchanged with a plurality of server devices in a server pool to obtain signal data regarding a plurality of signals associated with the network traffic. A determination is made when there is a sever health anomaly for one or more of the server devices based on an application of a server health prediction model to the signal data. The server health prediction model includes a plurality of predictive health targets each based at least in part on historical signal data for one or more of the signals and having an associated threshold value. A mitigation action is initiated when the determining indicates there is a sever health anomaly for one or more of the server devices.

This technology has a number of associated advantages including providing methods, non-transitory computer readable media, and health analysis apparatuses that more effectively analyze server health. This technology facilitates monitoring of server health based on observed network traffic signal data and the application of dynamically generated models to the signal data. The models generated by this technology consider historical server health anomalies and associated signal data and are more effective to predict future server health anomalies. Additionally, this technology advantageously leverages a more effective server health analysis to make load balancing decisions, thereby improving resource utilization, round trip times to client devices, and the associated end user experience.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of a network environment with an exemplary health analysis apparatus with a traffic management computing device and an analytic server computing device;

FIG. 2 is a block diagram of the exemplary traffic management computing device shown in FIG. 1;

FIG. 3 is a block diagram of the exemplary analytic server computing device shown in FIG. 2;

FIG. 4 is a flowchart of an exemplary method for analyzing server health; and

FIG. 5 is an exemplary connection lifetime timing diagram.

DETAILED DESCRIPTION

Referring to FIG. 1, a system diagram of an exemplary network environment 10 including an exemplary health analysis apparatus 12 with a traffic management computing device 14 and an analytic server computing device 16 is illustrated. In this example, the traffic management computing device 14 is coupled to client devices 18(1)-18(n) through communication network(s) 20, server devices 22(1)-22(m) through a local area network (LAN) 24, and the analytic server computing device 16 through a direct connection, although the traffic management computing device 12, client devices 18(1)-18(n), server devices 22(1)-22(m), and analytic server computing device 16 may be coupled together via other topologies. The network environment 10 may also include other network devices, such as one or more routers and/or switches, for example, which are well known in the art and thus will not be described herein. This technology provides a number of advantages including methods, non-transitory computer readable media, and health analysis apparatuses that facilitate more effective dynamic health monitoring of server devices and load balancing of network traffic to the server devices.

Referring to FIGS. 1-2, the traffic management computing device 14 may perform any number of functions in addition to analyzing server health including detecting and sending signal data to the analytic server computing device 16, and accelerating and optimizing network traffic communicated between the server devices 22(1)-22(m) and client devices 18(1)-18(n) in order to improve the user experience. The traffic management computing device 14 can also perform other functions on the network traffic, such as load balancing the network traffic to the server devices 22(1)-22(m) and/or implementing firewalls or other security measures on behalf of web applications executing on the server devices, for example. The traffic management computing device 14 includes processor(s) 26, a memory 28, and a communication interface 30, which are coupled together by a bus 32 or other communication link, although the traffic management computing device 14 may include other types and numbers of elements in other configurations.

The processor(s) 26 of the traffic management computing device 14 may execute programmed instructions for any number of the functions identified above and/or described herein for dynamic health monitoring of server devices 22(1)-22(m) and load balancing network traffic among the server devices 22(1)-22(m) and, optionally, managing network traffic and/or optimizing service of content requests, for example. The processor(s) 26 of the traffic management computing device 14 may include one or more central processing units and/or general purpose processors with one or more processing cores, for example, although other types of processor(s) can also be used.

The memory 28 of the traffic management computing device 14 stores these programmed instructions for one or more aspects of the present technology as described and illustrated herein, although some or all of the programmed instructions could be stored and executed elsewhere. A variety of different types of memory storage devices, such as random access memory (RAM), read only memory (ROM), flash, hard disk drives, solid state drives, or other computer readable medium which is read from and written to by a magnetic, optical, or other reading and writing system that is coupled to the processor(s) 26, can be used for the memory 28.

Accordingly, the memory 28 of the traffic management computing device 14 can store one or more applications that can include computer executable instructions that, when executed by the traffic management computing device 14, cause the traffic management computing device 14 to perform actions, such as to transmit, receive, or otherwise process messages, for example, and to perform other actions described and illustrated below with reference to FIGS. 4-5. The application(s) can be implemented as modules or components of other applications. Further, the application(s) can be implemented as operating system extensions, module, plugins, or the like.

Even further, the application(s) may be operative in a cloud-based computing environment. The application(s) can be executed within or as virtual machine(s) or virtual server(s) that may be managed in a cloud-based computing environment. Also, the application(s), and even the traffic management computing device 14 itself, may be located in virtual server(s) running in a cloud-based computing environment rather than being tied to one or more specific physical network computing devices. Also, the application(s) may be running in one or more virtual machines (VMs) executing on the traffic management computing device 14. Additionally, in one or more embodiments of this technology, virtual machine(s) running on the traffic management computing device 14 may be managed or supervised by a hypervisor.

In this particular example, the memory 28 further includes a load balancing module 34 that includes a load balancing policy 36 and a server health observer module 38. The load balancing module 34 is configured to implement the load balancing policy 36 to distribute network traffic among the server devices 22(1)-22(m) based on a server health score for each of the server devices 22(1)-22(m), as described and illustrated in more detail later. The server health observer module 38 is configured to monitor server-side signals to obtain data and apply models in order to determine the health of the server devices 22(1)-22(m) or whether one or more of the server devices 22(1)-22(m) is experiencing an anomaly representative of a health issue, also as described and illustrated in more detail later.

The communication interface 30 of the traffic management computing device 14 operatively couples and communicates between the traffic management computing device 14, client devices 18(1)-18(n), server devices 22(1)-22(m), and analytic server computing device 16, which are all coupled together by the LAN 24, communication network(s) 20 and direct connection(s), although other types and numbers of communication networks or systems with other types and numbers of connections and configurations to other devices and elements. By way of example only, the LAN and communication network(s) 20 can use TCP/IP over Ethernet and industry-standard protocols, including NFS, CIFS, SOAP, XML, LDAP, and SNMP, although other types and numbers of communication networks, can be used.

Referring to FIGS. 1-3, the analytic server computing device 16 is configured to process signal data received from the traffic management computing device 14 and generate server health models that facilitate subsequent network traffic analysis by the traffic management computing device 14. While the analytic server computing device 16 is illustrated in this particular example as a separate device that allows significant processing related to model generation and machine learning to be performed separate from the traffic management computing device 14, in other examples, the functionality provided by the analytic server computing device 16 can be performed by analytic modules included in the memory 28 of the traffic management computing device 14, and other configurations can also be used.

In yet other examples, the analytic server computing device 16 can be located in a local network or outside of a local network and accessible via a cloud architecture, for example. In this particular example, the analytic server computing device 16 includes processor(s) 40, a memory 42, and a communication interface 44, which are coupled together by a bus 46 or other communication link, although the analytic server computing device 16 may include other types and numbers of elements in other configurations.

The processor(s) 40 of the analytic server computing device 16 may execute programmed instructions for any number of the functions identified above and/or described herein for generating server health models that facilitate the identification of anomalous network traffic by the traffic management computing device 14. The processor(s) 40 of the analytic server computing device 16 may include one or more central processing units and/or general purpose processors with one or more processing cores, for example.

The memory 42 of the analytic server computing device 16 stores these programmed instructions for one or more aspects of the present technology as described and illustrated herein, although some or all of the programmed instructions could be stored and executed elsewhere. A variety of different types of memory storage devices, such as random access memory (RAM), read only memory (ROM), flash memory, hard disk drives, solid state drives, or other computer readable medium which is read from and written to by a magnetic, optical, or other reading and writing system that is coupled to the processor(s) 40, can be used for the memory 42.

Accordingly, the memory 42 of the analytic server computing device 16 can store one or more applications that can include computer executable instructions that, when executed by the analytic server computing device 16, cause the analytic server computing device 16 to perform actions, such as to transmit, receive, or otherwise process messages, for example, and to perform other actions described and illustrated below with reference to FIGS. 4-5. The application(s) can be implemented as modules or components of other applications. Further, the application(s) can be implemented as operating system extensions, module, plugins, or the like.

Even further, the application(s) may be operative in a cloud-based computing environment. The application(s) can be executed within or as virtual machine(s) or virtual server(s) that may be managed in a cloud-based computing environment. Also, the application(s), and even the analytic server computing device 20 itself, may be located in virtual server(s) running in a cloud-based computing environment rather than being tied to one or more specific physical network computing devices. Also, the application(s) may be running in one or more virtual machines (VMs) executing on the analytic server computing device 16. Additionally, in one or more embodiments of this technology, virtual machine(s) running on the analytic server computing device 16 may be managed or supervised by a hypervisor.

In this particular example, the memory 42 of the analytic server computing device 16 further includes a model builder module 48. The model builder module 48 in this example is configured to dynamically generate models that can be applied by the traffic management computing device 14 to identify anomalous network traffic impacting the health of one or more of the server devices 22(1)-22(m), although the model builder module 48 can be configured to execute other types and/or numbers of other health monitoring functions and/or other operations. The models can include thresholds for any number of signals and an estimated number of pending requests that, when one or more are exceeded, indicate that an anomaly exists, as described and illustrated in more detail later.

The communication interface 44 of the analytic server computing device 16 operatively couples and communicates with the traffic management computing device 14, which is coupled to the analytic server computing devices by a direct connection or LAN (not shown), although other types and numbers of communication networks or systems with other types and numbers of connections and configurations to other devices and elements can also be used.

Each of the server devices 22(1)-22(m) in this example includes one or more processors, a memory, and a communication interface, which are coupled together by a bus or other communication link, although other numbers and types of network devices could be used. The server devices 22(1)-22(m) in this example process requests received from the client devices 18(1)-18(n) via the communication network(s) 20 according to the HTTP-based application RFC protocol, for example. Various applications may be operating on the server devices 22(1)-22(m) and transmitting data (e.g., files or Web pages) to the client devices 18(1)-18(n) via the traffic management computing device 14 in response to requests from the client devices 18(1)-18(n). The server devices 22(1)-22(m) may be hardware or software or may represent a system with multiple servers in a pool, which may include internal or external networks.

Although the server devices 22(1)-22(m) are illustrated as single devices, one or more actions of each of the server devices 22(1)-22(m) may be distributed across one or more distinct network computing devices that together comprise one or more of the server devices 22(1)-22(m). Moreover, the server devices 22(1)-22(m) are not limited to a particular configuration. Thus, the server devices 22(1)-22(m) may contain a plurality of network computing devices that operate using a master/slave approach, whereby one of the network computing devices of the server devices 22(1)-22(m) operate to manage and/or otherwise coordinate operations of the other network computing devices. The server devices 22(1)-22(m) may operate as a plurality of network computing devices within a cluster architecture, a peer-to peer architecture, virtual machines, or within a cloud architecture, for example.

Thus, the technology disclosed herein is not to be construed as being limited to a single environment and other configurations and architectures are also envisaged. For example, one or more of the server devices 22(1)-22(m) can operate within the traffic management computing devices 14 itself, rather than as a stand-alone server device. In this example, the one or more of the server devices 22(1)-22(m) operate within the memory 28 of the traffic management computing devices 14.

The client devices 18(1)-18(n) in this example include any type of computing device that can generate, receive, and process network traffic, such as mobile computing devices, desktop computing devices, laptop computing devices, tablet computing devices, virtual machines (including cloud-based computers), or the like. Each of the client devices 18(1)-18(n) in this example includes a processor, a memory, and a communication interface, which are coupled together by a bus or other communication link, although other numbers and types of network devices could be used.

The client devices 18(1)-18(n) may run interface applications, such as standard web browsers or standalone client applications, that may provide an interface to make requests for, and receive content stored on, one or more of the server devices 22(1)-22(m) via the communication network(s) 20. The client devices 18(1)-18(n) may further include a display device, such as a display screen or touchscreen, and/or an input device, such as a keyboard for example.

Although the exemplary network environment with the traffic management computing device 14, client devices 18(1)-18(n), server devices 22(1)-22(m), analytic server computing device 16, LAN 24, and communication network(s) 20 are described and illustrated herein, other types and numbers of systems, devices, components, and elements in other topologies can be used. It is to be understood that the systems of the examples described herein are for exemplary purposes, as many variations of the specific hardware and software used to implement the examples are possible, as will be appreciated by those skilled in the relevant art(s).

One or more of the components depicted in the network environment 10, such as the traffic management computing device 14, client devices 18(1)-18(n), server devices 22(1)-22(m), or analytic server computing device 16, for example, may be configured to operate as virtual instances on the same physical machine. In other words, one or more of the traffic management computing device 14, client devices 18(1)-18(n), server devices 22(1)-22(m), or analytic server computing device 16 may operate on the same physical device rather than as separate devices communicating through communication network(s). Additionally, there may be more or fewer traffic management computing device, client devices, server devices, or analytic server computing devices than illustrated in FIG. 1.

In addition, two or more computing systems or devices can be substituted for any one of the systems or devices in any example. Accordingly, principles and advantages of distributed processing, such as redundancy and replication also can be implemented, as desired, to increase the robustness and performance of the devices and systems of the examples. The examples may also be implemented on computer system(s) that extend across any suitable network using any suitable interface mechanisms and traffic technologies, including by way of example only teletraffic in any suitable form (e.g., voice and modem), wireless traffic media, wireless traffic networks, cellular traffic networks, G3 traffic networks, Public Switched Telephone Network (PSTNs), Packet Data Networks (PDNs), the Internet, intranets, and combinations thereof.

The examples may also be embodied as non-transitory computer readable media having instructions stored thereon for one or more aspects of the present technology as described and illustrated by way of the examples herein, as described herein, which when executed by a processor, cause the processor to carry out the steps necessary to implement the methods of the examples, as described and illustrated herein.

Exemplary methods for analyzing server health will now be described with reference to FIGS. 1-5. Referring more specifically to FIG. 3, a method for monitoring the health of the server devices 22(1)-22(m) is illustrated. In step 400 in this example, the health analysis apparatus 12 monitors network traffic exchanged with the server devices 22(1)-22(m) to obtain raw health signal data.

Optionally, the traffic management computing device 14 of the health analysis apparatus 12 can obtain the signal data periodically (e.g., every second) over a specified period of time, and can combine the signal data into a snapshot, which is sent to the model builder module 48 of the analytic server computing device 14. The signal data can include transactions per second, requests per second, request jitter, response jitter, drops per second, pending transactions per second, concurrent connections per second, bytes in per second, bytes out per second, upload time, download time, idle time, inter-packet time, mean and/or median latency over a specified time period, server errors per second, or redirected requests per second, for example, although other signals can also be used.

Referring more specifically to FIG. 5, an exemplary connection lifetime timing diagram is illustrated. In this example, a request is received by the traffic management computing device 14 followed by an upload time. Subsequent to the upload, the connection experiences a latency associated with one of the server devices 22(1)-22(m) to which the request is directed. In response to the request, the one of the server devices 22(1)-22(m) provides requested data during a download time. Subsequent to the download time, the traffic management computing device 14 is idle during an idle time before and after a subsequent request is received. In this example, one or more of the upload time, server latency, download time, inter-packet time, or idle time can be recorded as part of the signal data.

Referring back to FIG. 4, in step 402, the health analysis apparatus 12 determines whether there is a predicted anomaly in the network traffic monitored in step 400. In order to determine whether there is a predicted anomaly, the traffic management computing device 12 can apply a server health prediction model received from the model builder module 48 of the analytic server computing device 16 to the signal data obtained for each of the server devices 22(1)-22(m) in step 400. The server health prediction model can be dynamically modified and can result from a machine learning process applied to historical predicted and actual anomalies and associated historical signal data.

In this particular example, the server health prediction model can include a plurality of predictive health targets, such as expressions or values that have a benign range and for which associated boundaries or thresholds have been determined over time based on server health feedback. The predictive health targets can be saturation expressions, pending transaction (or pending request) expressions, and/or ratios that have components corresponding to the signals for which signal data is obtained in step 400 (e.g., pending transactions to concurrent connections or request jitter for the traffic management computing device 14 to request jitter for one or more of the server devices 22(1)-22(m)), although other types of expressions can also be used for the predictive health targets. Additionally, the predictive health targets can include expected latency, expected drops per second, expected server errors per second, or redirected requests per second, for example, and values for other signals can also be used for the predictive health targets.

The predictive health targets are generated using both current value(s) of the predictive health target component(s) in the signal data (e.g., current number of pending requests) and also historical signal data for the component(s) in the predictive health targets that is obtained in one or more other time windows. Accordingly, in this example, predictive health target and signal data values during time interval of W seconds (in which there are W values for every predictive health target and every signal data value (assuming signal data is collected every second)) and predictive health target values and signal data values during the following W seconds (i.e., in time t+w) are used in one or more machine learning algorithms to build the server health prediction model(s) that include the predictive health targets analyzed in step 402.

Accordingly, the health analysis apparatus 12 can predict a server health anomaly in step 402 when a specified number or one or more specified ones of the threshold values for one or more of the predictive health targets in the server health prediction model have been exceeded, for example. In other examples, other methods of determining whether there is a predicted server health anomaly using the signal data and predictive health targets also can be used. If the health analysis apparatus 12 determines there is a server health anomaly, then the Yes branch is taken to step 404.

In step 404, the health analysis apparatus 12 initiates a mitigation action. The mitigation action can include blocking network traffic having certain characteristics, intentionally dropping packets from certain of the client devices 18(1)-18(n), presenting certain client devices with challenges before proceeding to accept network traffic in order to confirm whether the client devices 18(1)-18(n) are malicious bots, redirecting network traffic to one or more relatively healthy of the server devices 22(1)-22(m), or any other type of mitigation action intended to defend the one or more of the server devices. Accordingly, the mitigation action can be initiated by the traffic management computing device 14, for example. Subsequent to initiating the mitigation action, or if the health analysis apparatus 12 determines in step 402 that there is not a predicted anomaly in the network traffic and the No branch is taken, then the health analysis apparatus 12 proceeds to step 406.

In step 406, the health analysis apparatus 12 generates a server health score for each of the server devices 22(1)-22(m) based on an application of the server health prediction model. The server health score corresponds with an estimated saturation level of each of the server devices 22(1)-22(m) and can be generated by the model builder module 48 of the analytic server computing device 16, for example. The server devices 22(1)-22(m) are considered saturated when additional requests or connections cannot be serviced. Accordingly the server health score represents an estimated saturation level or how close one of the server devices 22(1)-22(m) is to saturation.

In this particular example, the server health score can be generated based on the value of one or more of the expressions or values corresponding to one or more of the predictive health targets as compared to the threshold for the one or more expressions or values corresponding to the one or more of the predictive health targets, although server health scores can be generated based on other values and data. The predictive health target thresholds are optionally dynamic and determined over time based on historical values of the various signals when anomalies or health issues have occurred with each of the server devices 22(1)-22(m).

In step 408, the health analysis apparatus 12 updates the load balancing policy 36 based on the generated server health scores. Optionally, the load balancing module 34 of the traffic management computing device 14 can store one or more load balancing policies (e.g., load balancing policy 36) that can be applied to received network traffic in order to identify ones of the server devices 22(1)-22(m) to which the network traffic should be sent in order to provide a relatively balanced distribution of the network traffic across the server pool.

By using the server health scores as the load balancing criterion, ones of the server devices 22(1)-22(m) that are relatively healthy can process more of the network traffic received by the traffic management computing device 14, thereby resulting in more efficient responses to requests sent by the client devices 18(1)-18(n) and an associated improved user experience. Optionally, the server health scores can also be used to dynamically spawn an additional server device to join the pool and share the load of the server devices 22(1)-22(m) or to shift the load to a relatively healthy one or more other of the server devices 22(1)-22(m), for example.

Subsequent to updating the load balancing policy, the health analysis apparatus 12 proceeds back to step 400 and continues monitoring network traffic exchanged with the server devices 22(1)-22(m) to obtain signal data. In some examples, a plurality of iterations of steps 400-408 illustrated in FIG. 4 are performed by the health analysis apparatus 12, so that the predictive health targets of the server health prediction model become relatively reliable, prior to mitigation actions being taken in step 404 and/or the load balancing policy 36 being updated in step 408. During the iterative process, and based on the machine learning performed by the analytic server computing device 16, the predictive health targets of the server health prediction model can be validated and invalidated based on actual health target indicators representative of actual health issues or anomalies with one or more of the server devices 22(1)-22(m).

Accordingly, with this technology, server health issues can advantageously be identified and mitigated earlier and before service significantly deteriorates. Additionally, threshold values for various signals can be learned and dynamically changed over time based on historical signal data and anomalies to provide a relatively effective model for predicting future anomalies with respect to server health. Further, network traffic can be more effectively load balanced to servers in a server pool based on the health of the server.

Having thus described the basic concept of the disclosed technology, it will be rather apparent to those skilled in the art that the foregoing detailed disclosure is intended to be presented by way of example only, and is not limiting. Various alterations, improvements, and modifications will occur and are intended to those skilled in the art, though not expressly stated herein. These alterations, improvements, and modifications are intended to be suggested hereby, and are within the spirit and scope of the disclosed technology. Additionally, the recited order of processing elements or sequences, or the use of numbers, letters, or other designations therefore, is not intended to limit the claimed processes to any order except as may be specified in the claims. Accordingly, the disclosed technology is limited only by the following claims and equivalents thereto.

Claims

1. A method for analyzing server health, comprising: monitoring network traffic exchanged with a plurality of servers to obtain signal data regarding a plurality of signals associated with the network traffic;applying a server health prediction model to the signal data to generate a server health score for each of the servers, the server health prediction model comprising a plurality of predictive health targets each based at least in part on historical signal data and having an associated threshold value, the historical signal data obtained based on an analysis of monitored network traffic exchanged between one or more of a plurality of clients and one or more of the servers; andupdating a stored load balancing policy based on the generated server health score for each of the servers to modify a distribution across the servers of network traffic received from the clients.

2. The method of claim 1, wherein the server health score for each of the servers, which is generated based the application of the server health prediction model to the signal data obtained based on the monitoring of network traffic exchanged with the servers, comprises an estimated saturation level of each of the servers.

3. The method of claim 1, further comprising: determining when there is a predicted server health anomaly based on the application of the server health prediction model to the signal data; andinitiating a mitigation action when the determination indicates that there is a predicted server health anomaly.

4. The method of claim 1, wherein the signal data comprises one or more of transactions per second, requests per second, request jitter, response jitter, drops per second, pending transactions per second, bytes in per second, bytes out per second, upload time, download time, idle time, inter-packet time, or response codes.

5. The method of claim 1, further comprising determining when a specified number or one or more specified ones of the threshold values has been exceeded.

6. A non-transitory computer readable medium having stored thereon instructions for analyzing server health comprising executable code that, when executed by one or more processors, causes the processors to: monitor network traffic exchanged with a plurality of servers to obtain signal data regarding a plurality of signals associated with the network traffic;apply a server health prediction model to the signal data to generate a server health score for each of the servers, the server health prediction model comprising a plurality of predictive health targets each based at least in part on historical signal data and having an associated threshold value, the historical signal data obtained based on an analysis of monitored network traffic exchanged between one or more of a plurality of clients and one or more of the servers; andupdate a stored load balancing policy based on the generated server health score for each of the servers to modify a distribution across the servers of network traffic received from the clients.

7. The non-transitory computer readable medium of claim 6, wherein the server health score for each of the servers, which is generated based the application of the server health prediction model to the signal data obtained based on the monitoring of network traffic exchanged with the servers, comprises an estimated saturation level of each of the servers.

8. The non-transitory computer readable medium of claim 6, wherein the executable code, when executed by the processors, further causes the processors to: determine when there is a predicted server health anomaly based on the application of the server health prediction model to the signal data; andinitiate a mitigation action when the determination indicates that there is a predicted server health anomaly.

9. The non-transitory computer readable medium of claim 6, wherein the signal data comprises one or more of transactions per second, requests per second, request jitter, response jitter, drops per second, pending transactions per second, bytes in per second, bytes out per second, upload time, download time, idle time, inter-packet time, or response codes.

10. The non-transitory computer readable medium of claim 6, wherein the executable code, when executed by the processors, further causes the processors to determine when a specified number or one or more specified ones of the threshold values has been exceeded.

11. A health analysis apparatus, comprising memory comprising programmed instructions stored thereon and one or more processors configured to be capable of executing the stored programmed instructions to: monitor network traffic exchanged with a plurality of servers to obtain signal data regarding a plurality of signals associated with the network traffic;apply a server health prediction model to the signal data to generate a server health score for each of the servers, the server health prediction model comprising a plurality of predictive health targets each based at least in part on historical signal data and having an associated threshold value, the historical signal data obtained based on an analysis of monitored network traffic exchanged between one or more of a plurality of clients and one or more of the servers; andupdate a stored load balancing policy based on the generated server health score for each of the servers to modify a distribution across the servers of network traffic received from the clients.

12. The health analysis apparatus of claim 11, wherein the server health score for each of the servers, which is generated based the application of the server health prediction model to the signal data obtained based on the monitoring of network traffic exchanged with the servers, comprises an estimated saturation level of each of the servers.

13. The health analysis apparatus of claim 11, wherein the processors are further configured to be capable of executing the stored programmed instructions to: determine when there is a predicted server health anomaly based on the application of the server health prediction model to the signal data; andinitiate a mitigation action when the determination indicates that there is a predicted server health anomaly.

14. The health analysis apparatus of claim 11, wherein the signal data comprises one or more of transactions per second, requests per second, request jitter, response jitter, drops per second, pending transactions per second, bytes in per second, bytes out per second, upload time, download time, idle time, inter-packet time, or response codes.

15. The health analysis apparatus of claim 11, wherein the processors are further configured to be capable of executing the stored programmed instructions to determine when a specified number or one or more specified ones of the threshold values has been exceeded.

16. A network traffic management system, comprising one or more health analysis apparatuses, a plurality of client devices, and a plurality of server devices, the network traffic management system comprising memory comprising programmed instructions stored thereon and one or more processors configured to be capable of executing the stored programmed instructions to: monitor network traffic exchanged with the server devices to obtain signal data regarding a plurality of signals associated with the network traffic;apply a server health prediction model to the signal data to generate a server health score for each of the server devices, the server health prediction model comprising a plurality of predictive health targets each based at least in part on historical signal data and having an associated threshold value, the historical signal data obtained based on an analysis of monitored network traffic exchanged between one or more of the client devices and one or more of the server devices; andupdate a stored load balancing policy based on the generated server health score for each of the server devices to modify a distribution across the server devices of network traffic received from the client devices.

17. The network traffic management system of claim 16, wherein the server health score for each of the server devices, which is generated based the application of the server health prediction model to the signal data obtained based on the monitoring of network traffic exchanged with the server devices, comprises an estimated saturation level of each of the server devices.

18. The network traffic management system of claim 16, wherein the processors are further configured to be capable of executing the stored programmed instructions to: determine when there is a predicted server health anomaly based on the application of the server health prediction model to the signal data; andinitiate a mitigation action when the determination indicates that there is a predicted server health anomaly.

19. The network traffic management system of claim 16, wherein the signal data comprises one or more of transactions per second, requests per second, request jitter, response jitter, drops per second, pending transactions per second, bytes in per second, bytes out per second, upload time, download time, idle time, inter-packet time, or response codes.

20. The network traffic management system of claim 16, wherein the processors are further configured to be capable of executing the stored programmed instructions to determine when a specified number or one or more specified ones of the threshold values has been exceeded.