Predictika · Application delivery appliances

An AI sales agent's BIG-IP appliance order, checked before it goes out

An AI sales agent prepares an order for BIG-IP iSeries appliances: the platform, the number of units and whether they run active-active or as an active-standby pair, the traffic and SSL load the customer needs, the ports, optics and power supplies on each unit, and options such as hardware compression, TurboFlex, vCMP guests and FIPS. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the figures F5 publishes in its BIG-IP System data sheet and returns a verdict, every reason and the figures it computed.

Scope: BIG-IP iSeries appliances (i2600 to i15800): published throughput, RSA SSL, ports, optics, power supplies, vCMP, hardware compression, TurboFlex and FIPS rows from the data sheet. See everything this validator covers

This validator models an order for one BIG-IP iSeries platform at a time, based on the figures in the public F5 BIG-IP System data sheet, not just the scenarios shown below.

  • Published platform rows — fourteen platforms (the 2600, 2800, 4600, 4800, 5600, 5800, 7600, 7800, 10600, 10800, 11600, 11800, 15600 and 15800). Each row gives layer 7 throughput, RSA SSL transactions a second (2K keys), the 10 GE SFP+, 40 GE QSFP+ and 100 GE QSFP28 port counts, the maximum number of vCMP guests, and whether it lists hardware compression, TurboFlex profiles, a FIPS variant and one or two power supplies.
  • Capacity — layer 7 throughput and RSA SSL across the units that carry traffic against what the customer needs (an active-standby pair counts one unit as carrying traffic), and the throughput the quote states, which must equal the computed one. A requirement above 80 percent of the throughput goes to a person for approval; that threshold is a guideline set by this model, not an F5 figure.
  • Ports, optics and power — the ports needed on each unit against the ports the model has, optics that must be from F5, one or two power supplies, and DC power, which the data sheet lists as a pair of supplies.
  • Options — hardware compression and TurboFlex (listed for the 800 series only), vCMP guests (a per-model maximum, none on the 600 series), FIPS 140-2 Level 3 SSL (listed for the i5800, i7800 and i15800 only, and not available to vCMP guests) and the 40 degree Celsius operating limit.

Any BIG-IP iSeries appliance order that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (F5 does not publish list prices, so this page shows no price build-up), software modules and licensing tiers (LTM, Advanced WAF, AFM, APM and others), on-demand upgrades between the 600 and 800 series, the 10000v and 10350v models, other BIG-IP platforms such as rSeries, VELOS and VIPRION, virtual editions, and support terms. The data sheet is dated 2022, and F5 has other current platforms.

The AI sales agentBuilds the appliance order from the conversation and says it is ready.
Predictika validatorChecks every rule at once and returns the verdict, each finding, and the computed figures.

7 scenarios in 3 groups · each group runs from the agent’s first attempt to the outcome · click one to switch

What the agent proposes

The AI sales agent says

Illustrative agent statement. The validator receives the structured request below.

Proposing agent: AI sales agent (scripted)

Before sending, the agent must call Recorded run

Predictika validator


      

        
        
        
      

      
Waiting for the call.

      
    

  

  
  

    
Where rules-based CPQ gets it wrong
Click to expand
    
Our assessment of what a typical rules-based CPQ (product rules, price rules and validation rules, as in Salesforce CPQ) does with the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.

    
What the validator caughtTypical rules-based CPQPredictika
Only the platform rows the data sheet lists (14 iSeries platforms)Only with workarounds
Each platform has its own throughput, SSL, port, vCMP and power figures. A rules-based CPQ needs a compatibility table or one rule per row, and every data sheet update means editing them.		The platform arrives in one request and is matched to the published row on every call.
Totals across units, with an active-standby pair counting one unitOnly with care
Capacity is a per-unit figure times the units that carry traffic, which depends on the high availability mode, and then it is compared with two different requirements.		The units that carry traffic, the totals and both comparisons are computed on every call.
Options that exist only on some platformsOnly with care
Hardware compression, TurboFlex, vCMP guests and FIPS each apply to a different set of models, and FIPS also excludes vCMP guests. Each is a product rule that has to be kept in step.		Platform and every option are in one request and every availability rule is checked together.
Throughput the quote statesOnly with care
The stated figure is derived from the chosen row and the units, so a rule has to recompute it and compare.		The stated throughput is a field in the request, compared with the computed one on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote is saved or a step is completed, so a person often fixes one error and meets the next.		All findings come back together, whatever order the choices arrived in.

  

  

    
What this validator catches: findings from its test cases
Click to expand
    
Every rule is checked on every call, in any order the choices arrive; one problem never hides another. Blocking findings stop the quote; approval findings need a person to sign off; everything else is computed.

    
1The quote has 0 units; at least 1 is needed.
2The quote needs 60 Gbps of layer 7 throughput; the units that carry traffic give 0 Gbps in total.
3The quote needs 50,000 RSA (2K key) SSL transactions a second; the units that carry traffic give 0 in total.
4An active-standby pair needs at least 2 units; the quote has 1.
5The quote needs 30 Gbps of layer 7 throughput; the units that carry traffic give 0 Gbps in total.
6The quote needs 30,000 RSA (2K key) SSL transactions a second; the units that carry traffic give 0 in total.
7The quote needs 100 Gbps of layer 7 throughput; the units that carry traffic give 80 Gbps in total.
8The quote needs 70 Gbps of layer 7 throughput against 80 Gbps from the units that carry traffic, above the 80 percent headroom this model uses as a guideline; a person should confirm the sizing.
9The quote needs 90,000 RSA (2K key) SSL transactions a second; the units that carry traffic give 80,000 in total.
10The quote needs 9 10 GE SFP+ ports on each unit; the selected model has 8.
11The quote needs 5 40 GE QSFP+ ports on each unit; the selected model has 4.
12The data sheet lists no 100 GE QSFP28 ports on the selected model; the quote needs 1 on each unit.
13The data sheet lists no 10 GE SFP+ ports on the selected model; the quote needs 2 on each unit.
14The data sheet lists no 40 GE QSFP+ ports on the selected model; the quote needs 1 on each unit.
15The data sheet lists hardware compression only for the 800 series models; the selected model has software compression only.
16The data sheet lists TurboFlex performance profiles only for the 800 series models; the selected model has none.
17The quote needs 13 vCMP guests; the selected model supports at most 12.
18The data sheet lists no vCMP guests for the selected model; the quote needs 1.
19FIPS 140-2 Level 3 SSL is listed only for the i5800 (i5820-DF), i7800 (i7820-DF) and i15,800 (i15,820-DF) platforms.
20The data sheet notes that vCMP guest access to FIPS resources is not supported.
21The selected model ships with two power supplies; a single-supply quote is not listed.
22The DC power option is listed as a pair of supplies; the quote has 1.

  

  

    
Scenarios and results
Click to expand
    
The recorded result of the validator for each scenario on this page: the request the AI agent sent, the verdict, and every finding, in the order they appear above.

    
1. Compression, TurboFlex and vCMP: Three i10600 appliances in an active-standby set with compression, TurboFlex and vCMP guests

    
The agent's request. Please quote 3 BIG-IP i10600 appliances in active-standby mode for a data center that needs 180 Gbps of layer 7 throughput and 70,000 RSA SSL transactions a second. Each unit needs 8 10 GE SFP+ ports, 4 40 GE QSFP+ ports and 0 100 GE QSFP28 ports, with 2 AC power supplies. The order also needs hardware compression and TurboFlex. It needs 6 vCMP guests. The optics are from F5 and the room is at 25 degrees Celsius. The quote states 160 Gbps of layer 7 throughput in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Layer 7 throughput below the requirement. The quote needs 180 Gbps of layer 7 throughput; the units that carry traffic give 160 Gbps in total.
    • 
      
  • Hardware compression needs an 800 series model. The data sheet lists hardware compression only for the 800 series models; the selected model has software compression only.
    • 
      
  • TurboFlex needs an 800 series model. The data sheet lists TurboFlex performance profiles only for the 800 series models; the selected model has none.
    • 
      
  • Model has no vCMP guests. The data sheet lists no vCMP guests for the selected model; the quote needs 6.
    • 
    

    
2. Compression, TurboFlex and vCMP: The agent resubmits after fixing the errors

    
The agent's request. Please quote 4 BIG-IP i10800 appliances in active-standby mode for a data center that needs 180 Gbps of layer 7 throughput and 70,000 RSA SSL transactions a second. Each unit needs 8 10 GE SFP+ ports, 4 40 GE QSFP+ ports and 0 100 GE QSFP28 ports, with 2 AC power supplies. The order also needs hardware compression and TurboFlex. It needs 6 vCMP guests. The optics are from F5 and the room is at 25 degrees Celsius. The quote states 240 Gbps of layer 7 throughput in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 3 units carry traffic and give 240 Gbps of layer 7 throughput in total, against 180 Gbps required.
    • 
    

    
3. Ports and power supplies: Two i7800 appliances with FIPS, 100 GE ports and a single power supply

    
The agent's request. Please quote 2 BIG-IP i7800 appliances in active-active mode for a data center that needs 60 Gbps of layer 7 throughput and 50,000 RSA SSL transactions a second. Each unit needs 0 10 GE SFP+ ports, 2 40 GE QSFP+ ports and 2 100 GE QSFP28 ports, with 1 AC power supply. The order also needs hardware compression and FIPS 140-2 Level 3 SSL. The optics are from F5 and the room is at 25 degrees Celsius. The quote states 80 Gbps of layer 7 throughput in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Model has no 100 GE QSFP28 ports. The data sheet lists no 100 GE QSFP28 ports on the selected model; the quote needs 2 on each unit.
    • 
      
  • Model ships with two power supplies. The selected model ships with two power supplies; a single-supply quote is not listed.
    • 
    

    
4. Ports and power supplies: The agent resubmits after fixing the errors

    
The agent's request. Please quote 2 BIG-IP i15800 appliances in active-active mode for a data center that needs 60 Gbps of layer 7 throughput and 50,000 RSA SSL transactions a second. Each unit needs 0 10 GE SFP+ ports, 2 40 GE QSFP+ ports and 2 100 GE QSFP28 ports, with 2 AC power supplies. The order also needs hardware compression and FIPS 140-2 Level 3 SSL. The optics are from F5 and the room is at 25 degrees Celsius. The quote states 320 Gbps of layer 7 throughput in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 2 units carry traffic and give 320 Gbps of layer 7 throughput in total, against 60 Gbps required.
    • 
    

    
5. Optics, room temperature and headroom: An active-standby pair of i5800 appliances with third-party optics in a warm room

    
The agent's request. Please quote 2 BIG-IP i5800 appliances in active-standby mode for a data center that needs 35 Gbps of layer 7 throughput and 30,000 RSA SSL transactions a second. Each unit needs 4 10 GE SFP+ ports, 2 40 GE QSFP+ ports and 0 100 GE QSFP28 ports, with 2 DC power supplies. The optics are from another vendor and the room is at 42 degrees Celsius. The quote states 70 Gbps of layer 7 throughput in total. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Optics must be from F5. The data sheet states that only optics provided by F5 are supported.
    • 
      
  • Ambient temperature above the operating range. The room temperature of 42 degrees Celsius is above the 40 degree operating limit.
    • 
      
  • Stated throughput does not match. The quote states 70 Gbps of layer 7 throughput; the units that carry traffic give 35 Gbps.
    • 
      
  • Little throughput headroom. The quote needs 35 Gbps of layer 7 throughput against 35 Gbps from the units that carry traffic, above the 80 percent headroom this model uses as a guideline; a person should confirm the sizing.
    • 
    

    
6. Optics, room temperature and headroom: The agent resubmits after fixing the errors

    
The agent's request. Please quote 2 BIG-IP i5800 appliances in active-standby mode for a data center that needs 35 Gbps of layer 7 throughput and 30,000 RSA SSL transactions a second. Each unit needs 4 10 GE SFP+ ports, 2 40 GE QSFP+ ports and 0 100 GE QSFP28 ports, with 2 DC power supplies. The optics are from F5 and the room is at 35 degrees Celsius. The quote states 35 Gbps of layer 7 throughput in total. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Little throughput headroom. The quote needs 35 Gbps of layer 7 throughput against 35 Gbps from the units that carry traffic, above the 80 percent headroom this model uses as a guideline; a person should confirm the sizing.
    • 
      
  • Order summary. 1 unit carry traffic and give 35 Gbps of layer 7 throughput in total, against 35 Gbps required.
    • 
    

    
7. Optics, room temperature and headroom: The agent resubmits after fixing the errors

    
The agent's request. Please quote 3 BIG-IP i5800 appliances in active-standby mode for a data center that needs 35 Gbps of layer 7 throughput and 30,000 RSA SSL transactions a second. Each unit needs 4 10 GE SFP+ ports, 2 40 GE QSFP+ ports and 0 100 GE QSFP28 ports, with 2 DC power supplies. The optics are from F5 and the room is at 35 degrees Celsius. The quote states 70 Gbps of layer 7 throughput in total. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 2 units carry traffic and give 70 Gbps of layer 7 throughput in total, against 35 Gbps required.
    • 
    

  

  

    
Narration transcript
Click to expand
    
The text of the narrated walkthrough, step by step.

    
    
      
  1. Welcome to a demo of Predictika's Validator. An AI sales agent prepares an order for BIG-IP iSeries appliances: the platform, the number of units and whether they run active-active or as an active-standby pair, the traffic and SSL load the customer needs, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the figures F5 publishes in its BIG-IP System data sheet and returns a verdict, every reason and the figures it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote 3 BIG-IP i10600 appliances in active-standby mode for a data center that needs 180 Gbps of layer 7 throughput and 70,000 RSA SSL transactions a second. Each unit needs 8 10 GE SFP+ ports, 4 40 GE QSFP+ ports and 0 100 GE QSFP28 ports, with 2 AC power supplies. The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with four findings, all in one answer. The quote needs 180 Gbps of layer 7 throughput; the units that carry traffic give 160 Gbps in total. The data sheet lists hardware compression only for the 800 series models; the selected model has software compression only. The data sheet lists TurboFlex performance profiles only for the 800 series models; the selected model has none. The other one is on screen.
    6. 
      
  6. The validator returned four blocking findings in one answer. A check that reads one field at a time would pass most of this; it shows only when the choices in the quote are checked against each other.
    7. 
      
  7. What this shows: for throughput the quote states, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the model to BIG-IP i10800, the units to 4 and the stated total layer 7 throughput as to 240 Gbps.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All four earlier findings are cleared, and the validator's figure matches the quote.
    13. 
      
  13. Four problems caught in one call, fixed in one pass, and a clean quote. That is the value of Predictika's validator: catching mistakes before they reach the customer.
    14.